[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-aws-remediating-with-aws-security-agent":3,"mdc-659w1a-key":35,"related-org-aws-remediating-with-aws-security-agent":1820,"related-repo-aws-remediating-with-aws-security-agent":1989},{"slug":4,"name":4,"fn":5,"description":6,"org":7,"tags":11,"stars":24,"repoUrl":25,"updatedAt":26,"license":27,"forks":28,"topics":29,"repo":30,"sourceUrl":33,"mdContent":34},"remediating-with-aws-security-agent","remediate AWS Security Agent findings","Pull AWS Security Agent findings (penetration tests and code reviews) and drive remediation. Use this whenever the user mentions Security Agent, security findings, pentest or penetration test results, code review findings, vulnerabilities found in their AWS account, \"what did the security scan find\", remediating or triaging security risks, or wants to start fixing reported vulnerabilities — even if they don't name the service explicitly. Trigger it for phrases like \"get my security findings\", \"what vulnerabilities do we have\", \"let's fix the pentest results\", or \"triage the security report\". The skill discovers scans, exports findings to a gitignored local directory (so sensitive exploit detail is never committed), produces a prioritized triage summary, and offers to start fixing the highest-risk issues.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},"aws","AWS (Amazon)","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Faws.png",[12,16,18,21],{"name":13,"slug":14,"type":15},"Security","security","tag",{"name":17,"slug":8,"type":15},"AWS",{"name":19,"slug":20,"type":15},"Code Analysis","code-analysis",{"name":22,"slug":23,"type":15},"Debugging","debugging",1822,"https:\u002F\u002Fgithub.com\u002Faws\u002Fagent-toolkit-for-aws","2026-07-12T08:44:12.011083",null,157,[],{"repoUrl":25,"stars":24,"forks":28,"topics":31,"description":32},[],"Official, AWS-supported MCP servers, skills, and plugins to help AI agents build on AWS","https:\u002F\u002Fgithub.com\u002Faws\u002Fagent-toolkit-for-aws\u002Ftree\u002FHEAD\u002Fplugins\u002Faws-agents-for-devsecops\u002Fskills\u002Fremediating-with-aws-security-agent","---\nname: remediating-with-aws-security-agent\ndescription: >-\n  Pull AWS Security Agent findings (penetration tests and code reviews) and drive\n  remediation. Use this whenever the user mentions Security Agent, security findings,\n  pentest or penetration test results, code review findings, vulnerabilities found in\n  their AWS account, \"what did the security scan find\", remediating or triaging security\n  risks, or wants to start fixing reported vulnerabilities — even if they don't name the\n  service explicitly. Trigger it for phrases like \"get my security findings\", \"what\n  vulnerabilities do we have\", \"let's fix the pentest results\", or \"triage the security\n  report\". The skill discovers scans, exports findings to a gitignored local directory\n  (so sensitive exploit detail is never committed), produces a prioritized triage\n  summary, and offers to start fixing the highest-risk issues.\n---\n\n# Security Agent Remediation\n\nAWS Security Agent is a frontier agent that runs on-demand penetration tests and code\nreviews against a customer's applications and reports verified security risks. This skill\ntakes you from \"I have findings somewhere in AWS\" to \"I'm actively fixing the most\nimportant ones,\" while keeping the sensitive exploit detail out of source control.\n\nThe flow has four stages, and they matter in order:\n\n1. **Discover** which scans exist and how the account is configured (live, read-only).\n2. **Export** the findings to a local gitignored directory.\n3. **Triage** the findings into a prioritized, human-readable plan.\n4. **Remediate** by offering to fix the highest-risk issues.\n\n## Why the ordering and the guardrails matter\n\nFindings contain working attack scripts, reproduction steps, file paths, and sometimes\nleaked secrets or environment details. If that lands in a Git repo, a customer can\naccidentally commit and publish a step-by-step exploit for their own production system.\nSo the non-negotiable rule is: **findings are written only to `.security-agent\u002F`, and that\npath is gitignored before anything is written.**\n\n## Stage 1: Discover scans (live, read-only)\n\nFind out what the account has. All commands are read-only `list-*` operations.\n\nAWS Security Agent organizes data as a hierarchy — work down it:\n\n```\nApplication (account + Region)\n└── Agent Space        (workspace for design review, code review, and pentests)\n    ├── Penetration test → Pentest job → Findings\n    └── Code review      → Code review job → Findings\n```\n\nRun these to orient yourself and show the user what exists:\n\n```bash\naws securityagent list-agent-spaces\naws securityagent list-pentests          --agent-space-id \u003Cas-...>\naws securityagent list-code-reviews      --agent-space-id \u003Cas-...>\naws securityagent list-pentest-jobs-for-pentest         --agent-space-id \u003Cas-...> --pentest-id \u003Cpt-...>\naws securityagent list-code-review-jobs-for-code-review --agent-space-id \u003Cas-...> --code-review-id \u003Ccr-...>\n```\n\nJob `status` is one of `IN_PROGRESS`, `STOPPING`, `STOPPED`, `FAILED`, `COMPLETED`. Only\n`COMPLETED` jobs have a stable, full set of findings.\n\n### Match the codebase to a scan, then confirm\n\nAgent spaces, pentests, and code reviews are named after the application they target.\nBefore asking the user to pick from a raw list, make an informed guess about which scan\ncorresponds to *this* repository — the user is working in a codebase for a reason, and\nthe relevant findings are almost always for the app in front of them.\n\nInfer the app identity from the workspace using cheap, high-signal sources:\n\n- The repository \u002F root directory name and the Git remote URL (`git remote -v`).\n- Project manifests and their `name`\u002F`description` (`package.json`, `pyproject.toml`,\n  `*.csproj`, `go.mod`, `Cargo.toml`).\n- README titles, product\u002Fsteering docs, and any obvious product or company name.\n- Distinctive frameworks or domains that match a scan title.\n\nCompare those signals against the agent space \u002F scan names (case-insensitive, allow\npartial and fuzzy matches).\nThen **always confirm before exporting** — present your best guess and your reasoning, and\nlet the user correct it:\n\n> \"This repo looks like **`\u003Cproduct>`** (from `\u003Csignal>`), which matches the **\u003Cname>** agent\n> space. Use that, or pick another? [Other Agent Space names, ...]\"\n\nIf nothing matches with reasonable confidence, say so plainly and show the full list rather\nthan forcing a wrong guess. Never export from a guessed scan without the user's confirmation.\n\n## Stage 2: Export findings to `.security-agent\u002F` (gitignored)\n\nPull findings using AWS CLI commands. Write everything into `.security-agent\u002F` in the repo —\nnever to chat or stdout — because findings include working attack scripts, reproduction\nsteps, and sometimes leaked secrets.\n\n### 1. Lock down the output directory before pulling anything\n\n```bash\nmkdir -p .security-agent\necho '*' > .security-agent\u002F.gitignore\n```\n\n### 2. Resolve the latest COMPLETED job\n\nYou should already have the `agentSpaceId` and the pentest\u002Fcode-review id from Stage 1.\nList jobs for the chosen scan:\n\n```bash\n# Pentest jobs:\naws securityagent list-pentest-jobs-for-pentest \\\n  --agent-space-id \u003Cas-...> --pentest-id \u003Cpt-...>\n\n# Code review jobs:\naws securityagent list-code-review-jobs-for-code-review \\\n  --agent-space-id \u003Cas-...> --code-review-id \u003Ccr-...>\n```\n\nPaginate by passing `--next-token` from the previous response until absent. Filter the\njob summaries to `status == \"COMPLETED\"`. If none are COMPLETED, stop and tell the user\n\"No completed jobs found. Please wait for a job to complete or check job statuses.\"\nOtherwise, pick the COMPLETED job with the greatest `createdAt` timestamp.\n\n### 3. List finding summaries and filter by confidence\n\n```bash\n# Pentest findings:\naws securityagent list-findings \\\n  --agent-space-id \u003Cas-...> --pentest-job-id \u003Cpj-...>\n\n# Code review findings:\naws securityagent list-findings \\\n  --agent-space-id \u003Cas-...> --code-review-job-id \u003Ccj-...>\n```\n\nPaginate on `--next-token` until exhausted. Confidence values from weakest to strongest:\n`FALSE_POSITIVE`, `UNCONFIRMED`, `LOW`, `MEDIUM`, `HIGH`.\n**Keep only `HIGH` and `MEDIUM` by default.** Widen only when the user explicitly asks.\n\n### 4. Fetch full detail in batches of 25\n\n`batch-get-findings` accepts at most 25 ids per call. Chunk the filtered finding ids into\ngroups of 25:\n\n```bash\naws securityagent batch-get-findings \\\n  --agent-space-id \u003Cas-...> \\\n  --finding-ids \u003Cfid-1> \u003Cfid-2> ... \u003Cfid-25>\n```\n\nTag each returned finding with its source (`pentest` or `code-review`) before writing,\nso triage in Stage 3 can tell them apart.\n\n### 5. Write findings into `.security-agent\u002F`\n\nGroup findings by job id. For each job, write a full markdown report to\n`.security-agent\u002Ffindings_\u003CjobId>.md` with ALL fields returned by the API (findingId,\nname, description, riskLevel, riskType, confidence, status, codeLocations, remediationCode,\nand any other fields). Do not leave off any fields.\n\n### Edge cases\n\n- **No agent space, scan, or COMPLETED job** — stop and surface that to the user rather\n  than retrying.\n- **Credentials or service unavailable** — confirm with `aws sts get-caller-identity` and\n  check the Region (default `us-east-1`; Security Agent is regional).\n- **Don't paste finding contents into chat** beyond short titles and counts. The detail\n  belongs in the gitignored files.\n\n## Stage 3: Triage into a prioritized plan\n\nRank by risk, because remediation time is finite and a CRITICAL unauthenticated RCE\noutranks a LOW informational finding every time. Read the exported `findings_*.md`\nfiles from `.security-agent\u002F` and sort them deterministically.\n\n### Ranking rules\n\nSort ascending by this composite key (lower wins, i.e. more urgent first):\n\n1. **Risk level**, in this order:\n   `CRITICAL` (0) → `HIGH` (1) → `MEDIUM` (2) → `LOW` (3) → `INFORMATIONAL` (4) →\n   `UNKNOWN` \u002F missing (5).\n2. **Risk score**, highest first. `riskScore` is a numeric string on pentest findings\n   (e.g. `\"10.0\"`), often absent on code-review findings — treat missing as the lowest\n   possible score so it sorts after scored findings of the same level.\n3. **Confidence**, in this order:\n   `HIGH` (0) → `MEDIUM` (1) → `LOW` (2) → `UNCONFIRMED` (3) → `FALSE_POSITIVE` (4).\n\nAlso compute a severity-count summary across all findings (e.g. `2 CRITICAL · 5 HIGH ·\n3 MEDIUM`) for the header of the report.\n\n### Pulling the code location\n\nFor each finding, derive a single short `location` string:\n\n- If `filePath` is set, use it as-is.\n- Otherwise, take `codeLocations[0]`. Strip the scanner's sandbox prefix from `filePath`\n  (everything up to and including that marker) so the path is repo-relative; if that\n  marker isn't present, fall back to the basename. Append `:\u003ClineStart>` when present.\n- If neither is available (typical for some pentest findings), leave it blank and\n  describe the affected endpoint or attack chain in the impact line instead.\n\n### Summary format\n\nWrite a compact summary for the user:\n\n```\n## Security Agent triage — \u003Cagent space name>\n\n\u003CN> findings exported (\u003CP pentest, C code review>) · confidence: \u003Clevels> · severity: \u003Ccounts>\n\n### Priority order\n1. [CRITICAL · score 10.0 · HIGH confidence] \u003Cfinding name>\n   - Type: \u003CriskType> · Source: \u003Cpentest|code-review>\n   - Where: \u003Cfile:line or endpoint, if present>\n   - Impact: \u003Cone-line plain-language summary>\n2. [HIGH · ...] ...\n\n### Recommended remediation order\n\u003Cshort rationale: which to fix first and why — e.g. \"1 and 3 are both\nunauthenticated RCE on internet-facing endpoints; fix those before the\nstored-XSS issues.\">\n```\n\nIf more than ~10 findings, show the top N in detail and summarize the rest as a count\nby severity at the bottom.\n\n### What to keep out of chat\n\nThe full `description`, `reasoning`, and `attackScript` stay in the gitignored files —\nthey contain working exploit detail. In the chat summary keep impact lines to one line\neach, in plain language. Code-review findings usually carry a `filePath`\u002Flocation and a\n`suggestedFix`; call those out since they map directly to repo changes. Pentest findings\ndescribe endpoints and attack chains; map them to the responsible code where you can.\nLook for findings that corroborate each other (a pentest and a code review flagging the\nsame root cause) — those are strong signals for what to fix first.\n\n## Stage 4: Offer to remediate\n\nAfter presenting the triage, offer to start fixing — don't silently begin editing code.\n\nAsk the user something like: \"Want me to start fixing the top finding(s)? I'd recommend\nstarting with #1 (\u003Cname>).\" If they agree, work top-down by priority:\n\n1. Read the finding detail from the gitignored export file (location, description, suggested fix).\n2. Open the affected file and apply the fix via the editor.\n3. Report one line per fix: \"Fixed {name} in `{filePath}:{lineStart}`.\"\n\nIf the user wants to handle several findings, fix one at a time (or one cluster of related\nfindings) so each change stays reviewable, and proceed in the priority order from Stage 3.\n\n## Notes and edge cases\n\n- **No completed jobs**: a scan may still be `IN_PROGRESS`. Tell the user; offer to re-check\n  later rather than exporting a partial job.\n- **Re-running**: each run overwrites the files for that job id. The directory is safe to\n  delete; it only holds exported copies, not source-of-truth data.\n- **Multiple accounts\u002FRegions**: findings are Region-scoped. If the user expected results\n  and got none, confirm the region matches where Security Agent is configured.\n- **Data handling**: treat exported findings as sensitive. They are copies of verified\n  exploits against the user's own systems.\n",{"data":36,"body":37},{"name":4,"description":6},{"type":38,"children":39},"root",[40,49,55,60,106,113,132,138,151,156,168,173,407,464,471,484,489,573,585,628,633,646,658,664,723,729,742,903,932,938,1097,1166,1172,1183,1291,1312,1323,1336,1342,1391,1397,1417,1423,1428,1552,1565,1571,1584,1633,1639,1644,1653,1658,1664,1706,1712,1717,1727,1753,1758,1764,1814],{"type":41,"tag":42,"props":43,"children":45},"element","h1",{"id":44},"security-agent-remediation",[46],{"type":47,"value":48},"text","Security Agent Remediation",{"type":41,"tag":50,"props":51,"children":52},"p",{},[53],{"type":47,"value":54},"AWS Security Agent is a frontier agent that runs on-demand penetration tests and code\nreviews against a customer's applications and reports verified security risks. This skill\ntakes you from \"I have findings somewhere in AWS\" to \"I'm actively fixing the most\nimportant ones,\" while keeping the sensitive exploit detail out of source control.",{"type":41,"tag":50,"props":56,"children":57},{},[58],{"type":47,"value":59},"The flow has four stages, and they matter in order:",{"type":41,"tag":61,"props":62,"children":63},"ol",{},[64,76,86,96],{"type":41,"tag":65,"props":66,"children":67},"li",{},[68,74],{"type":41,"tag":69,"props":70,"children":71},"strong",{},[72],{"type":47,"value":73},"Discover",{"type":47,"value":75}," which scans exist and how the account is configured (live, read-only).",{"type":41,"tag":65,"props":77,"children":78},{},[79,84],{"type":41,"tag":69,"props":80,"children":81},{},[82],{"type":47,"value":83},"Export",{"type":47,"value":85}," the findings to a local gitignored directory.",{"type":41,"tag":65,"props":87,"children":88},{},[89,94],{"type":41,"tag":69,"props":90,"children":91},{},[92],{"type":47,"value":93},"Triage",{"type":47,"value":95}," the findings into a prioritized, human-readable plan.",{"type":41,"tag":65,"props":97,"children":98},{},[99,104],{"type":41,"tag":69,"props":100,"children":101},{},[102],{"type":47,"value":103},"Remediate",{"type":47,"value":105}," by offering to fix the highest-risk issues.",{"type":41,"tag":107,"props":108,"children":110},"h2",{"id":109},"why-the-ordering-and-the-guardrails-matter",[111],{"type":47,"value":112},"Why the ordering and the guardrails matter",{"type":41,"tag":50,"props":114,"children":115},{},[116,118],{"type":47,"value":117},"Findings contain working attack scripts, reproduction steps, file paths, and sometimes\nleaked secrets or environment details. If that lands in a Git repo, a customer can\naccidentally commit and publish a step-by-step exploit for their own production system.\nSo the non-negotiable rule is: ",{"type":41,"tag":69,"props":119,"children":120},{},[121,123,130],{"type":47,"value":122},"findings are written only to ",{"type":41,"tag":124,"props":125,"children":127},"code",{"className":126},[],[128],{"type":47,"value":129},".security-agent\u002F",{"type":47,"value":131},", and that\npath is gitignored before anything is written.",{"type":41,"tag":107,"props":133,"children":135},{"id":134},"stage-1-discover-scans-live-read-only",[136],{"type":47,"value":137},"Stage 1: Discover scans (live, read-only)",{"type":41,"tag":50,"props":139,"children":140},{},[141,143,149],{"type":47,"value":142},"Find out what the account has. All commands are read-only ",{"type":41,"tag":124,"props":144,"children":146},{"className":145},[],[147],{"type":47,"value":148},"list-*",{"type":47,"value":150}," operations.",{"type":41,"tag":50,"props":152,"children":153},{},[154],{"type":47,"value":155},"AWS Security Agent organizes data as a hierarchy — work down it:",{"type":41,"tag":157,"props":158,"children":162},"pre",{"className":159,"code":161,"language":47},[160],"language-text","Application (account + Region)\n└── Agent Space        (workspace for design review, code review, and pentests)\n    ├── Penetration test → Pentest job → Findings\n    └── Code review      → Code review job → Findings\n",[163],{"type":41,"tag":124,"props":164,"children":166},{"__ignoreMap":165},"",[167],{"type":47,"value":161},{"type":41,"tag":50,"props":169,"children":170},{},[171],{"type":47,"value":172},"Run these to orient yourself and show the user what exists:",{"type":41,"tag":157,"props":174,"children":178},{"className":175,"code":176,"language":177,"meta":165,"style":165},"language-bash shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","aws securityagent list-agent-spaces\naws securityagent list-pentests          --agent-space-id \u003Cas-...>\naws securityagent list-code-reviews      --agent-space-id \u003Cas-...>\naws securityagent list-pentest-jobs-for-pentest         --agent-space-id \u003Cas-...> --pentest-id \u003Cpt-...>\naws securityagent list-code-review-jobs-for-code-review --agent-space-id \u003Cas-...> --code-review-id \u003Ccr-...>\n","bash",[179],{"type":41,"tag":124,"props":180,"children":181},{"__ignoreMap":165},[182,204,248,286,347],{"type":41,"tag":183,"props":184,"children":187},"span",{"class":185,"line":186},"line",1,[188,193,199],{"type":41,"tag":183,"props":189,"children":191},{"style":190},"--shiki-light:#E2931D;--shiki-default:#FFCB6B;--shiki-dark:#FFCB6B",[192],{"type":47,"value":8},{"type":41,"tag":183,"props":194,"children":196},{"style":195},"--shiki-light:#91B859;--shiki-default:#C3E88D;--shiki-dark:#C3E88D",[197],{"type":47,"value":198}," securityagent",{"type":41,"tag":183,"props":200,"children":201},{"style":195},[202],{"type":47,"value":203}," list-agent-spaces\n",{"type":41,"tag":183,"props":205,"children":207},{"class":185,"line":206},2,[208,212,216,221,226,232,237,243],{"type":41,"tag":183,"props":209,"children":210},{"style":190},[211],{"type":47,"value":8},{"type":41,"tag":183,"props":213,"children":214},{"style":195},[215],{"type":47,"value":198},{"type":41,"tag":183,"props":217,"children":218},{"style":195},[219],{"type":47,"value":220}," list-pentests",{"type":41,"tag":183,"props":222,"children":223},{"style":195},[224],{"type":47,"value":225},"          --agent-space-id",{"type":41,"tag":183,"props":227,"children":229},{"style":228},"--shiki-light:#39ADB5;--shiki-default:#89DDFF;--shiki-dark:#89DDFF",[230],{"type":47,"value":231}," \u003C",{"type":41,"tag":183,"props":233,"children":234},{"style":195},[235],{"type":47,"value":236},"as-..",{"type":41,"tag":183,"props":238,"children":240},{"style":239},"--shiki-light:#90A4AE;--shiki-default:#EEFFFF;--shiki-dark:#BABED8",[241],{"type":47,"value":242},".",{"type":41,"tag":183,"props":244,"children":245},{"style":228},[246],{"type":47,"value":247},">\n",{"type":41,"tag":183,"props":249,"children":251},{"class":185,"line":250},3,[252,256,260,265,270,274,278,282],{"type":41,"tag":183,"props":253,"children":254},{"style":190},[255],{"type":47,"value":8},{"type":41,"tag":183,"props":257,"children":258},{"style":195},[259],{"type":47,"value":198},{"type":41,"tag":183,"props":261,"children":262},{"style":195},[263],{"type":47,"value":264}," list-code-reviews",{"type":41,"tag":183,"props":266,"children":267},{"style":195},[268],{"type":47,"value":269},"      --agent-space-id",{"type":41,"tag":183,"props":271,"children":272},{"style":228},[273],{"type":47,"value":231},{"type":41,"tag":183,"props":275,"children":276},{"style":195},[277],{"type":47,"value":236},{"type":41,"tag":183,"props":279,"children":280},{"style":239},[281],{"type":47,"value":242},{"type":41,"tag":183,"props":283,"children":284},{"style":228},[285],{"type":47,"value":247},{"type":41,"tag":183,"props":287,"children":289},{"class":185,"line":288},4,[290,294,298,303,308,312,316,320,325,330,334,339,343],{"type":41,"tag":183,"props":291,"children":292},{"style":190},[293],{"type":47,"value":8},{"type":41,"tag":183,"props":295,"children":296},{"style":195},[297],{"type":47,"value":198},{"type":41,"tag":183,"props":299,"children":300},{"style":195},[301],{"type":47,"value":302}," list-pentest-jobs-for-pentest",{"type":41,"tag":183,"props":304,"children":305},{"style":195},[306],{"type":47,"value":307},"         --agent-space-id",{"type":41,"tag":183,"props":309,"children":310},{"style":228},[311],{"type":47,"value":231},{"type":41,"tag":183,"props":313,"children":314},{"style":195},[315],{"type":47,"value":236},{"type":41,"tag":183,"props":317,"children":318},{"style":239},[319],{"type":47,"value":242},{"type":41,"tag":183,"props":321,"children":322},{"style":228},[323],{"type":47,"value":324},">",{"type":41,"tag":183,"props":326,"children":327},{"style":195},[328],{"type":47,"value":329}," --pentest-id",{"type":41,"tag":183,"props":331,"children":332},{"style":228},[333],{"type":47,"value":231},{"type":41,"tag":183,"props":335,"children":336},{"style":195},[337],{"type":47,"value":338},"pt-..",{"type":41,"tag":183,"props":340,"children":341},{"style":239},[342],{"type":47,"value":242},{"type":41,"tag":183,"props":344,"children":345},{"style":228},[346],{"type":47,"value":247},{"type":41,"tag":183,"props":348,"children":350},{"class":185,"line":349},5,[351,355,359,364,369,373,377,381,385,390,394,399,403],{"type":41,"tag":183,"props":352,"children":353},{"style":190},[354],{"type":47,"value":8},{"type":41,"tag":183,"props":356,"children":357},{"style":195},[358],{"type":47,"value":198},{"type":41,"tag":183,"props":360,"children":361},{"style":195},[362],{"type":47,"value":363}," list-code-review-jobs-for-code-review",{"type":41,"tag":183,"props":365,"children":366},{"style":195},[367],{"type":47,"value":368}," --agent-space-id",{"type":41,"tag":183,"props":370,"children":371},{"style":228},[372],{"type":47,"value":231},{"type":41,"tag":183,"props":374,"children":375},{"style":195},[376],{"type":47,"value":236},{"type":41,"tag":183,"props":378,"children":379},{"style":239},[380],{"type":47,"value":242},{"type":41,"tag":183,"props":382,"children":383},{"style":228},[384],{"type":47,"value":324},{"type":41,"tag":183,"props":386,"children":387},{"style":195},[388],{"type":47,"value":389}," --code-review-id",{"type":41,"tag":183,"props":391,"children":392},{"style":228},[393],{"type":47,"value":231},{"type":41,"tag":183,"props":395,"children":396},{"style":195},[397],{"type":47,"value":398},"cr-..",{"type":41,"tag":183,"props":400,"children":401},{"style":239},[402],{"type":47,"value":242},{"type":41,"tag":183,"props":404,"children":405},{"style":228},[406],{"type":47,"value":247},{"type":41,"tag":50,"props":408,"children":409},{},[410,412,418,420,426,428,434,435,441,442,448,449,455,457,462],{"type":47,"value":411},"Job ",{"type":41,"tag":124,"props":413,"children":415},{"className":414},[],[416],{"type":47,"value":417},"status",{"type":47,"value":419}," is one of ",{"type":41,"tag":124,"props":421,"children":423},{"className":422},[],[424],{"type":47,"value":425},"IN_PROGRESS",{"type":47,"value":427},", ",{"type":41,"tag":124,"props":429,"children":431},{"className":430},[],[432],{"type":47,"value":433},"STOPPING",{"type":47,"value":427},{"type":41,"tag":124,"props":436,"children":438},{"className":437},[],[439],{"type":47,"value":440},"STOPPED",{"type":47,"value":427},{"type":41,"tag":124,"props":443,"children":445},{"className":444},[],[446],{"type":47,"value":447},"FAILED",{"type":47,"value":427},{"type":41,"tag":124,"props":450,"children":452},{"className":451},[],[453],{"type":47,"value":454},"COMPLETED",{"type":47,"value":456},". Only\n",{"type":41,"tag":124,"props":458,"children":460},{"className":459},[],[461],{"type":47,"value":454},{"type":47,"value":463}," jobs have a stable, full set of findings.",{"type":41,"tag":465,"props":466,"children":468},"h3",{"id":467},"match-the-codebase-to-a-scan-then-confirm",[469],{"type":47,"value":470},"Match the codebase to a scan, then confirm",{"type":41,"tag":50,"props":472,"children":473},{},[474,476,482],{"type":47,"value":475},"Agent spaces, pentests, and code reviews are named after the application they target.\nBefore asking the user to pick from a raw list, make an informed guess about which scan\ncorresponds to ",{"type":41,"tag":477,"props":478,"children":479},"em",{},[480],{"type":47,"value":481},"this",{"type":47,"value":483}," repository — the user is working in a codebase for a reason, and\nthe relevant findings are almost always for the app in front of them.",{"type":41,"tag":50,"props":485,"children":486},{},[487],{"type":47,"value":488},"Infer the app identity from the workspace using cheap, high-signal sources:",{"type":41,"tag":490,"props":491,"children":492},"ul",{},[493,506,563,568],{"type":41,"tag":65,"props":494,"children":495},{},[496,498,504],{"type":47,"value":497},"The repository \u002F root directory name and the Git remote URL (",{"type":41,"tag":124,"props":499,"children":501},{"className":500},[],[502],{"type":47,"value":503},"git remote -v",{"type":47,"value":505},").",{"type":41,"tag":65,"props":507,"children":508},{},[509,511,517,519,525,527,533,534,540,542,548,549,555,556,562],{"type":47,"value":510},"Project manifests and their ",{"type":41,"tag":124,"props":512,"children":514},{"className":513},[],[515],{"type":47,"value":516},"name",{"type":47,"value":518},"\u002F",{"type":41,"tag":124,"props":520,"children":522},{"className":521},[],[523],{"type":47,"value":524},"description",{"type":47,"value":526}," (",{"type":41,"tag":124,"props":528,"children":530},{"className":529},[],[531],{"type":47,"value":532},"package.json",{"type":47,"value":427},{"type":41,"tag":124,"props":535,"children":537},{"className":536},[],[538],{"type":47,"value":539},"pyproject.toml",{"type":47,"value":541},",\n",{"type":41,"tag":124,"props":543,"children":545},{"className":544},[],[546],{"type":47,"value":547},"*.csproj",{"type":47,"value":427},{"type":41,"tag":124,"props":550,"children":552},{"className":551},[],[553],{"type":47,"value":554},"go.mod",{"type":47,"value":427},{"type":41,"tag":124,"props":557,"children":559},{"className":558},[],[560],{"type":47,"value":561},"Cargo.toml",{"type":47,"value":505},{"type":41,"tag":65,"props":564,"children":565},{},[566],{"type":47,"value":567},"README titles, product\u002Fsteering docs, and any obvious product or company name.",{"type":41,"tag":65,"props":569,"children":570},{},[571],{"type":47,"value":572},"Distinctive frameworks or domains that match a scan title.",{"type":41,"tag":50,"props":574,"children":575},{},[576,578,583],{"type":47,"value":577},"Compare those signals against the agent space \u002F scan names (case-insensitive, allow\npartial and fuzzy matches).\nThen ",{"type":41,"tag":69,"props":579,"children":580},{},[581],{"type":47,"value":582},"always confirm before exporting",{"type":47,"value":584}," — present your best guess and your reasoning, and\nlet the user correct it:",{"type":41,"tag":586,"props":587,"children":588},"blockquote",{},[589],{"type":41,"tag":50,"props":590,"children":591},{},[592,594,603,605,611,613,619,621,626],{"type":47,"value":593},"\"This repo looks like ",{"type":41,"tag":69,"props":595,"children":596},{},[597],{"type":41,"tag":124,"props":598,"children":600},{"className":599},[],[601],{"type":47,"value":602},"\u003Cproduct>",{"type":47,"value":604}," (from ",{"type":41,"tag":124,"props":606,"children":608},{"className":607},[],[609],{"type":47,"value":610},"\u003Csignal>",{"type":47,"value":612},"), which matches the ",{"type":41,"tag":69,"props":614,"children":615},{},[616],{"type":41,"tag":516,"props":617,"children":618},{},[],{"type":47,"value":620}," agent\nspace. Use that, or pick another? ",{"type":41,"tag":183,"props":622,"children":623},{},[624],{"type":47,"value":625},"Other Agent Space names, ...",{"type":47,"value":627},"\"",{"type":41,"tag":50,"props":629,"children":630},{},[631],{"type":47,"value":632},"If nothing matches with reasonable confidence, say so plainly and show the full list rather\nthan forcing a wrong guess. Never export from a guessed scan without the user's confirmation.",{"type":41,"tag":107,"props":634,"children":636},{"id":635},"stage-2-export-findings-to-security-agent-gitignored",[637,639,644],{"type":47,"value":638},"Stage 2: Export findings to ",{"type":41,"tag":124,"props":640,"children":642},{"className":641},[],[643],{"type":47,"value":129},{"type":47,"value":645}," (gitignored)",{"type":41,"tag":50,"props":647,"children":648},{},[649,651,656],{"type":47,"value":650},"Pull findings using AWS CLI commands. Write everything into ",{"type":41,"tag":124,"props":652,"children":654},{"className":653},[],[655],{"type":47,"value":129},{"type":47,"value":657}," in the repo —\nnever to chat or stdout — because findings include working attack scripts, reproduction\nsteps, and sometimes leaked secrets.",{"type":41,"tag":465,"props":659,"children":661},{"id":660},"_1-lock-down-the-output-directory-before-pulling-anything",[662],{"type":47,"value":663},"1. Lock down the output directory before pulling anything",{"type":41,"tag":157,"props":665,"children":667},{"className":175,"code":666,"language":177,"meta":165,"style":165},"mkdir -p .security-agent\necho '*' > .security-agent\u002F.gitignore\n",[668],{"type":41,"tag":124,"props":669,"children":670},{"__ignoreMap":165},[671,689],{"type":41,"tag":183,"props":672,"children":673},{"class":185,"line":186},[674,679,684],{"type":41,"tag":183,"props":675,"children":676},{"style":190},[677],{"type":47,"value":678},"mkdir",{"type":41,"tag":183,"props":680,"children":681},{"style":195},[682],{"type":47,"value":683}," -p",{"type":41,"tag":183,"props":685,"children":686},{"style":195},[687],{"type":47,"value":688}," .security-agent\n",{"type":41,"tag":183,"props":690,"children":691},{"class":185,"line":206},[692,698,703,708,713,718],{"type":41,"tag":183,"props":693,"children":695},{"style":694},"--shiki-light:#6182B8;--shiki-default:#82AAFF;--shiki-dark:#82AAFF",[696],{"type":47,"value":697},"echo",{"type":41,"tag":183,"props":699,"children":700},{"style":228},[701],{"type":47,"value":702}," '",{"type":41,"tag":183,"props":704,"children":705},{"style":195},[706],{"type":47,"value":707},"*",{"type":41,"tag":183,"props":709,"children":710},{"style":228},[711],{"type":47,"value":712},"'",{"type":41,"tag":183,"props":714,"children":715},{"style":228},[716],{"type":47,"value":717}," >",{"type":41,"tag":183,"props":719,"children":720},{"style":195},[721],{"type":47,"value":722}," .security-agent\u002F.gitignore\n",{"type":41,"tag":465,"props":724,"children":726},{"id":725},"_2-resolve-the-latest-completed-job",[727],{"type":47,"value":728},"2. Resolve the latest COMPLETED job",{"type":41,"tag":50,"props":730,"children":731},{},[732,734,740],{"type":47,"value":733},"You should already have the ",{"type":41,"tag":124,"props":735,"children":737},{"className":736},[],[738],{"type":47,"value":739},"agentSpaceId",{"type":47,"value":741}," and the pentest\u002Fcode-review id from Stage 1.\nList jobs for the chosen scan:",{"type":41,"tag":157,"props":743,"children":745},{"className":175,"code":744,"language":177,"meta":165,"style":165},"# Pentest jobs:\naws securityagent list-pentest-jobs-for-pentest \\\n  --agent-space-id \u003Cas-...> --pentest-id \u003Cpt-...>\n\n# Code review jobs:\naws securityagent list-code-review-jobs-for-code-review \\\n  --agent-space-id \u003Cas-...> --code-review-id \u003Ccr-...>\n",[746],{"type":41,"tag":124,"props":747,"children":748},{"__ignoreMap":165},[749,758,778,822,831,839,859],{"type":41,"tag":183,"props":750,"children":751},{"class":185,"line":186},[752],{"type":41,"tag":183,"props":753,"children":755},{"style":754},"--shiki-light:#90A4AE;--shiki-light-font-style:italic;--shiki-default:#546E7A;--shiki-default-font-style:italic;--shiki-dark:#676E95;--shiki-dark-font-style:italic",[756],{"type":47,"value":757},"# Pentest jobs:\n",{"type":41,"tag":183,"props":759,"children":760},{"class":185,"line":206},[761,765,769,773],{"type":41,"tag":183,"props":762,"children":763},{"style":190},[764],{"type":47,"value":8},{"type":41,"tag":183,"props":766,"children":767},{"style":195},[768],{"type":47,"value":198},{"type":41,"tag":183,"props":770,"children":771},{"style":195},[772],{"type":47,"value":302},{"type":41,"tag":183,"props":774,"children":775},{"style":239},[776],{"type":47,"value":777}," \\\n",{"type":41,"tag":183,"props":779,"children":780},{"class":185,"line":250},[781,786,790,794,798,802,806,810,814,818],{"type":41,"tag":183,"props":782,"children":783},{"style":195},[784],{"type":47,"value":785},"  --agent-space-id",{"type":41,"tag":183,"props":787,"children":788},{"style":228},[789],{"type":47,"value":231},{"type":41,"tag":183,"props":791,"children":792},{"style":195},[793],{"type":47,"value":236},{"type":41,"tag":183,"props":795,"children":796},{"style":239},[797],{"type":47,"value":242},{"type":41,"tag":183,"props":799,"children":800},{"style":228},[801],{"type":47,"value":324},{"type":41,"tag":183,"props":803,"children":804},{"style":195},[805],{"type":47,"value":329},{"type":41,"tag":183,"props":807,"children":808},{"style":228},[809],{"type":47,"value":231},{"type":41,"tag":183,"props":811,"children":812},{"style":195},[813],{"type":47,"value":338},{"type":41,"tag":183,"props":815,"children":816},{"style":239},[817],{"type":47,"value":242},{"type":41,"tag":183,"props":819,"children":820},{"style":228},[821],{"type":47,"value":247},{"type":41,"tag":183,"props":823,"children":824},{"class":185,"line":288},[825],{"type":41,"tag":183,"props":826,"children":828},{"emptyLinePlaceholder":827},true,[829],{"type":47,"value":830},"\n",{"type":41,"tag":183,"props":832,"children":833},{"class":185,"line":349},[834],{"type":41,"tag":183,"props":835,"children":836},{"style":754},[837],{"type":47,"value":838},"# Code review jobs:\n",{"type":41,"tag":183,"props":840,"children":842},{"class":185,"line":841},6,[843,847,851,855],{"type":41,"tag":183,"props":844,"children":845},{"style":190},[846],{"type":47,"value":8},{"type":41,"tag":183,"props":848,"children":849},{"style":195},[850],{"type":47,"value":198},{"type":41,"tag":183,"props":852,"children":853},{"style":195},[854],{"type":47,"value":363},{"type":41,"tag":183,"props":856,"children":857},{"style":239},[858],{"type":47,"value":777},{"type":41,"tag":183,"props":860,"children":862},{"class":185,"line":861},7,[863,867,871,875,879,883,887,891,895,899],{"type":41,"tag":183,"props":864,"children":865},{"style":195},[866],{"type":47,"value":785},{"type":41,"tag":183,"props":868,"children":869},{"style":228},[870],{"type":47,"value":231},{"type":41,"tag":183,"props":872,"children":873},{"style":195},[874],{"type":47,"value":236},{"type":41,"tag":183,"props":876,"children":877},{"style":239},[878],{"type":47,"value":242},{"type":41,"tag":183,"props":880,"children":881},{"style":228},[882],{"type":47,"value":324},{"type":41,"tag":183,"props":884,"children":885},{"style":195},[886],{"type":47,"value":389},{"type":41,"tag":183,"props":888,"children":889},{"style":228},[890],{"type":47,"value":231},{"type":41,"tag":183,"props":892,"children":893},{"style":195},[894],{"type":47,"value":398},{"type":41,"tag":183,"props":896,"children":897},{"style":239},[898],{"type":47,"value":242},{"type":41,"tag":183,"props":900,"children":901},{"style":228},[902],{"type":47,"value":247},{"type":41,"tag":50,"props":904,"children":905},{},[906,908,914,916,922,924,930],{"type":47,"value":907},"Paginate by passing ",{"type":41,"tag":124,"props":909,"children":911},{"className":910},[],[912],{"type":47,"value":913},"--next-token",{"type":47,"value":915}," from the previous response until absent. Filter the\njob summaries to ",{"type":41,"tag":124,"props":917,"children":919},{"className":918},[],[920],{"type":47,"value":921},"status == \"COMPLETED\"",{"type":47,"value":923},". If none are COMPLETED, stop and tell the user\n\"No completed jobs found. Please wait for a job to complete or check job statuses.\"\nOtherwise, pick the COMPLETED job with the greatest ",{"type":41,"tag":124,"props":925,"children":927},{"className":926},[],[928],{"type":47,"value":929},"createdAt",{"type":47,"value":931}," timestamp.",{"type":41,"tag":465,"props":933,"children":935},{"id":934},"_3-list-finding-summaries-and-filter-by-confidence",[936],{"type":47,"value":937},"3. List finding summaries and filter by confidence",{"type":41,"tag":157,"props":939,"children":941},{"className":175,"code":940,"language":177,"meta":165,"style":165},"# Pentest findings:\naws securityagent list-findings \\\n  --agent-space-id \u003Cas-...> --pentest-job-id \u003Cpj-...>\n\n# Code review findings:\naws securityagent list-findings \\\n  --agent-space-id \u003Cas-...> --code-review-job-id \u003Ccj-...>\n",[942],{"type":41,"tag":124,"props":943,"children":944},{"__ignoreMap":165},[945,953,973,1018,1025,1033,1052],{"type":41,"tag":183,"props":946,"children":947},{"class":185,"line":186},[948],{"type":41,"tag":183,"props":949,"children":950},{"style":754},[951],{"type":47,"value":952},"# Pentest findings:\n",{"type":41,"tag":183,"props":954,"children":955},{"class":185,"line":206},[956,960,964,969],{"type":41,"tag":183,"props":957,"children":958},{"style":190},[959],{"type":47,"value":8},{"type":41,"tag":183,"props":961,"children":962},{"style":195},[963],{"type":47,"value":198},{"type":41,"tag":183,"props":965,"children":966},{"style":195},[967],{"type":47,"value":968}," list-findings",{"type":41,"tag":183,"props":970,"children":971},{"style":239},[972],{"type":47,"value":777},{"type":41,"tag":183,"props":974,"children":975},{"class":185,"line":250},[976,980,984,988,992,996,1001,1005,1010,1014],{"type":41,"tag":183,"props":977,"children":978},{"style":195},[979],{"type":47,"value":785},{"type":41,"tag":183,"props":981,"children":982},{"style":228},[983],{"type":47,"value":231},{"type":41,"tag":183,"props":985,"children":986},{"style":195},[987],{"type":47,"value":236},{"type":41,"tag":183,"props":989,"children":990},{"style":239},[991],{"type":47,"value":242},{"type":41,"tag":183,"props":993,"children":994},{"style":228},[995],{"type":47,"value":324},{"type":41,"tag":183,"props":997,"children":998},{"style":195},[999],{"type":47,"value":1000}," --pentest-job-id",{"type":41,"tag":183,"props":1002,"children":1003},{"style":228},[1004],{"type":47,"value":231},{"type":41,"tag":183,"props":1006,"children":1007},{"style":195},[1008],{"type":47,"value":1009},"pj-..",{"type":41,"tag":183,"props":1011,"children":1012},{"style":239},[1013],{"type":47,"value":242},{"type":41,"tag":183,"props":1015,"children":1016},{"style":228},[1017],{"type":47,"value":247},{"type":41,"tag":183,"props":1019,"children":1020},{"class":185,"line":288},[1021],{"type":41,"tag":183,"props":1022,"children":1023},{"emptyLinePlaceholder":827},[1024],{"type":47,"value":830},{"type":41,"tag":183,"props":1026,"children":1027},{"class":185,"line":349},[1028],{"type":41,"tag":183,"props":1029,"children":1030},{"style":754},[1031],{"type":47,"value":1032},"# Code review findings:\n",{"type":41,"tag":183,"props":1034,"children":1035},{"class":185,"line":841},[1036,1040,1044,1048],{"type":41,"tag":183,"props":1037,"children":1038},{"style":190},[1039],{"type":47,"value":8},{"type":41,"tag":183,"props":1041,"children":1042},{"style":195},[1043],{"type":47,"value":198},{"type":41,"tag":183,"props":1045,"children":1046},{"style":195},[1047],{"type":47,"value":968},{"type":41,"tag":183,"props":1049,"children":1050},{"style":239},[1051],{"type":47,"value":777},{"type":41,"tag":183,"props":1053,"children":1054},{"class":185,"line":861},[1055,1059,1063,1067,1071,1075,1080,1084,1089,1093],{"type":41,"tag":183,"props":1056,"children":1057},{"style":195},[1058],{"type":47,"value":785},{"type":41,"tag":183,"props":1060,"children":1061},{"style":228},[1062],{"type":47,"value":231},{"type":41,"tag":183,"props":1064,"children":1065},{"style":195},[1066],{"type":47,"value":236},{"type":41,"tag":183,"props":1068,"children":1069},{"style":239},[1070],{"type":47,"value":242},{"type":41,"tag":183,"props":1072,"children":1073},{"style":228},[1074],{"type":47,"value":324},{"type":41,"tag":183,"props":1076,"children":1077},{"style":195},[1078],{"type":47,"value":1079}," --code-review-job-id",{"type":41,"tag":183,"props":1081,"children":1082},{"style":228},[1083],{"type":47,"value":231},{"type":41,"tag":183,"props":1085,"children":1086},{"style":195},[1087],{"type":47,"value":1088},"cj-..",{"type":41,"tag":183,"props":1090,"children":1091},{"style":239},[1092],{"type":47,"value":242},{"type":41,"tag":183,"props":1094,"children":1095},{"style":228},[1096],{"type":47,"value":247},{"type":41,"tag":50,"props":1098,"children":1099},{},[1100,1102,1107,1109,1115,1116,1122,1123,1129,1130,1136,1137,1143,1145,1164],{"type":47,"value":1101},"Paginate on ",{"type":41,"tag":124,"props":1103,"children":1105},{"className":1104},[],[1106],{"type":47,"value":913},{"type":47,"value":1108}," until exhausted. Confidence values from weakest to strongest:\n",{"type":41,"tag":124,"props":1110,"children":1112},{"className":1111},[],[1113],{"type":47,"value":1114},"FALSE_POSITIVE",{"type":47,"value":427},{"type":41,"tag":124,"props":1117,"children":1119},{"className":1118},[],[1120],{"type":47,"value":1121},"UNCONFIRMED",{"type":47,"value":427},{"type":41,"tag":124,"props":1124,"children":1126},{"className":1125},[],[1127],{"type":47,"value":1128},"LOW",{"type":47,"value":427},{"type":41,"tag":124,"props":1131,"children":1133},{"className":1132},[],[1134],{"type":47,"value":1135},"MEDIUM",{"type":47,"value":427},{"type":41,"tag":124,"props":1138,"children":1140},{"className":1139},[],[1141],{"type":47,"value":1142},"HIGH",{"type":47,"value":1144},".\n",{"type":41,"tag":69,"props":1146,"children":1147},{},[1148,1150,1155,1157,1162],{"type":47,"value":1149},"Keep only ",{"type":41,"tag":124,"props":1151,"children":1153},{"className":1152},[],[1154],{"type":47,"value":1142},{"type":47,"value":1156}," and ",{"type":41,"tag":124,"props":1158,"children":1160},{"className":1159},[],[1161],{"type":47,"value":1135},{"type":47,"value":1163}," by default.",{"type":47,"value":1165}," Widen only when the user explicitly asks.",{"type":41,"tag":465,"props":1167,"children":1169},{"id":1168},"_4-fetch-full-detail-in-batches-of-25",[1170],{"type":47,"value":1171},"4. Fetch full detail in batches of 25",{"type":41,"tag":50,"props":1173,"children":1174},{},[1175,1181],{"type":41,"tag":124,"props":1176,"children":1178},{"className":1177},[],[1179],{"type":47,"value":1180},"batch-get-findings",{"type":47,"value":1182}," accepts at most 25 ids per call. Chunk the filtered finding ids into\ngroups of 25:",{"type":41,"tag":157,"props":1184,"children":1186},{"className":175,"code":1185,"language":177,"meta":165,"style":165},"aws securityagent batch-get-findings \\\n  --agent-space-id \u003Cas-...> \\\n  --finding-ids \u003Cfid-1> \u003Cfid-2> ... \u003Cfid-25>\n",[1187],{"type":41,"tag":124,"props":1188,"children":1189},{"__ignoreMap":165},[1190,1210,1237],{"type":41,"tag":183,"props":1191,"children":1192},{"class":185,"line":186},[1193,1197,1201,1206],{"type":41,"tag":183,"props":1194,"children":1195},{"style":190},[1196],{"type":47,"value":8},{"type":41,"tag":183,"props":1198,"children":1199},{"style":195},[1200],{"type":47,"value":198},{"type":41,"tag":183,"props":1202,"children":1203},{"style":195},[1204],{"type":47,"value":1205}," batch-get-findings",{"type":41,"tag":183,"props":1207,"children":1208},{"style":239},[1209],{"type":47,"value":777},{"type":41,"tag":183,"props":1211,"children":1212},{"class":185,"line":206},[1213,1217,1221,1225,1229,1233],{"type":41,"tag":183,"props":1214,"children":1215},{"style":195},[1216],{"type":47,"value":785},{"type":41,"tag":183,"props":1218,"children":1219},{"style":228},[1220],{"type":47,"value":231},{"type":41,"tag":183,"props":1222,"children":1223},{"style":195},[1224],{"type":47,"value":236},{"type":41,"tag":183,"props":1226,"children":1227},{"style":239},[1228],{"type":47,"value":242},{"type":41,"tag":183,"props":1230,"children":1231},{"style":228},[1232],{"type":47,"value":324},{"type":41,"tag":183,"props":1234,"children":1235},{"style":239},[1236],{"type":47,"value":777},{"type":41,"tag":183,"props":1238,"children":1239},{"class":185,"line":250},[1240,1245,1249,1254,1259,1263,1267,1272,1277,1281,1286],{"type":41,"tag":183,"props":1241,"children":1242},{"style":195},[1243],{"type":47,"value":1244},"  --finding-ids",{"type":41,"tag":183,"props":1246,"children":1247},{"style":228},[1248],{"type":47,"value":231},{"type":41,"tag":183,"props":1250,"children":1251},{"style":195},[1252],{"type":47,"value":1253},"fid-",{"type":41,"tag":183,"props":1255,"children":1256},{"style":228},[1257],{"type":47,"value":1258},"1>",{"type":41,"tag":183,"props":1260,"children":1261},{"style":228},[1262],{"type":47,"value":231},{"type":41,"tag":183,"props":1264,"children":1265},{"style":195},[1266],{"type":47,"value":1253},{"type":41,"tag":183,"props":1268,"children":1269},{"style":228},[1270],{"type":47,"value":1271},"2>",{"type":41,"tag":183,"props":1273,"children":1274},{"style":195},[1275],{"type":47,"value":1276}," ...",{"type":41,"tag":183,"props":1278,"children":1279},{"style":228},[1280],{"type":47,"value":231},{"type":41,"tag":183,"props":1282,"children":1283},{"style":195},[1284],{"type":47,"value":1285},"fid-2",{"type":41,"tag":183,"props":1287,"children":1288},{"style":228},[1289],{"type":47,"value":1290},"5>\n",{"type":41,"tag":50,"props":1292,"children":1293},{},[1294,1296,1302,1304,1310],{"type":47,"value":1295},"Tag each returned finding with its source (",{"type":41,"tag":124,"props":1297,"children":1299},{"className":1298},[],[1300],{"type":47,"value":1301},"pentest",{"type":47,"value":1303}," or ",{"type":41,"tag":124,"props":1305,"children":1307},{"className":1306},[],[1308],{"type":47,"value":1309},"code-review",{"type":47,"value":1311},") before writing,\nso triage in Stage 3 can tell them apart.",{"type":41,"tag":465,"props":1313,"children":1315},{"id":1314},"_5-write-findings-into-security-agent",[1316,1318],{"type":47,"value":1317},"5. Write findings into ",{"type":41,"tag":124,"props":1319,"children":1321},{"className":1320},[],[1322],{"type":47,"value":129},{"type":41,"tag":50,"props":1324,"children":1325},{},[1326,1328,1334],{"type":47,"value":1327},"Group findings by job id. For each job, write a full markdown report to\n",{"type":41,"tag":124,"props":1329,"children":1331},{"className":1330},[],[1332],{"type":47,"value":1333},".security-agent\u002Ffindings_\u003CjobId>.md",{"type":47,"value":1335}," with ALL fields returned by the API (findingId,\nname, description, riskLevel, riskType, confidence, status, codeLocations, remediationCode,\nand any other fields). Do not leave off any fields.",{"type":41,"tag":465,"props":1337,"children":1339},{"id":1338},"edge-cases",[1340],{"type":47,"value":1341},"Edge cases",{"type":41,"tag":490,"props":1343,"children":1344},{},[1345,1355,1381],{"type":41,"tag":65,"props":1346,"children":1347},{},[1348,1353],{"type":41,"tag":69,"props":1349,"children":1350},{},[1351],{"type":47,"value":1352},"No agent space, scan, or COMPLETED job",{"type":47,"value":1354}," — stop and surface that to the user rather\nthan retrying.",{"type":41,"tag":65,"props":1356,"children":1357},{},[1358,1363,1365,1371,1373,1379],{"type":41,"tag":69,"props":1359,"children":1360},{},[1361],{"type":47,"value":1362},"Credentials or service unavailable",{"type":47,"value":1364}," — confirm with ",{"type":41,"tag":124,"props":1366,"children":1368},{"className":1367},[],[1369],{"type":47,"value":1370},"aws sts get-caller-identity",{"type":47,"value":1372}," and\ncheck the Region (default ",{"type":41,"tag":124,"props":1374,"children":1376},{"className":1375},[],[1377],{"type":47,"value":1378},"us-east-1",{"type":47,"value":1380},"; Security Agent is regional).",{"type":41,"tag":65,"props":1382,"children":1383},{},[1384,1389],{"type":41,"tag":69,"props":1385,"children":1386},{},[1387],{"type":47,"value":1388},"Don't paste finding contents into chat",{"type":47,"value":1390}," beyond short titles and counts. The detail\nbelongs in the gitignored files.",{"type":41,"tag":107,"props":1392,"children":1394},{"id":1393},"stage-3-triage-into-a-prioritized-plan",[1395],{"type":47,"value":1396},"Stage 3: Triage into a prioritized plan",{"type":41,"tag":50,"props":1398,"children":1399},{},[1400,1402,1408,1410,1415],{"type":47,"value":1401},"Rank by risk, because remediation time is finite and a CRITICAL unauthenticated RCE\noutranks a LOW informational finding every time. Read the exported ",{"type":41,"tag":124,"props":1403,"children":1405},{"className":1404},[],[1406],{"type":47,"value":1407},"findings_*.md",{"type":47,"value":1409},"\nfiles from ",{"type":41,"tag":124,"props":1411,"children":1413},{"className":1412},[],[1414],{"type":47,"value":129},{"type":47,"value":1416}," and sort them deterministically.",{"type":41,"tag":465,"props":1418,"children":1420},{"id":1419},"ranking-rules",[1421],{"type":47,"value":1422},"Ranking rules",{"type":41,"tag":50,"props":1424,"children":1425},{},[1426],{"type":47,"value":1427},"Sort ascending by this composite key (lower wins, i.e. more urgent first):",{"type":41,"tag":61,"props":1429,"children":1430},{},[1431,1486,1512],{"type":41,"tag":65,"props":1432,"children":1433},{},[1434,1439,1441,1447,1449,1454,1456,1461,1463,1468,1470,1476,1478,1484],{"type":41,"tag":69,"props":1435,"children":1436},{},[1437],{"type":47,"value":1438},"Risk level",{"type":47,"value":1440},", in this order:\n",{"type":41,"tag":124,"props":1442,"children":1444},{"className":1443},[],[1445],{"type":47,"value":1446},"CRITICAL",{"type":47,"value":1448}," (0) → ",{"type":41,"tag":124,"props":1450,"children":1452},{"className":1451},[],[1453],{"type":47,"value":1142},{"type":47,"value":1455}," (1) → ",{"type":41,"tag":124,"props":1457,"children":1459},{"className":1458},[],[1460],{"type":47,"value":1135},{"type":47,"value":1462}," (2) → ",{"type":41,"tag":124,"props":1464,"children":1466},{"className":1465},[],[1467],{"type":47,"value":1128},{"type":47,"value":1469}," (3) → ",{"type":41,"tag":124,"props":1471,"children":1473},{"className":1472},[],[1474],{"type":47,"value":1475},"INFORMATIONAL",{"type":47,"value":1477}," (4) →\n",{"type":41,"tag":124,"props":1479,"children":1481},{"className":1480},[],[1482],{"type":47,"value":1483},"UNKNOWN",{"type":47,"value":1485}," \u002F missing (5).",{"type":41,"tag":65,"props":1487,"children":1488},{},[1489,1494,1496,1502,1504,1510],{"type":41,"tag":69,"props":1490,"children":1491},{},[1492],{"type":47,"value":1493},"Risk score",{"type":47,"value":1495},", highest first. ",{"type":41,"tag":124,"props":1497,"children":1499},{"className":1498},[],[1500],{"type":47,"value":1501},"riskScore",{"type":47,"value":1503}," is a numeric string on pentest findings\n(e.g. ",{"type":41,"tag":124,"props":1505,"children":1507},{"className":1506},[],[1508],{"type":47,"value":1509},"\"10.0\"",{"type":47,"value":1511},"), often absent on code-review findings — treat missing as the lowest\npossible score so it sorts after scored findings of the same level.",{"type":41,"tag":65,"props":1513,"children":1514},{},[1515,1520,1521,1526,1527,1532,1533,1538,1539,1544,1545,1550],{"type":41,"tag":69,"props":1516,"children":1517},{},[1518],{"type":47,"value":1519},"Confidence",{"type":47,"value":1440},{"type":41,"tag":124,"props":1522,"children":1524},{"className":1523},[],[1525],{"type":47,"value":1142},{"type":47,"value":1448},{"type":41,"tag":124,"props":1528,"children":1530},{"className":1529},[],[1531],{"type":47,"value":1135},{"type":47,"value":1455},{"type":41,"tag":124,"props":1534,"children":1536},{"className":1535},[],[1537],{"type":47,"value":1128},{"type":47,"value":1462},{"type":41,"tag":124,"props":1540,"children":1542},{"className":1541},[],[1543],{"type":47,"value":1121},{"type":47,"value":1469},{"type":41,"tag":124,"props":1546,"children":1548},{"className":1547},[],[1549],{"type":47,"value":1114},{"type":47,"value":1551}," (4).",{"type":41,"tag":50,"props":1553,"children":1554},{},[1555,1557,1563],{"type":47,"value":1556},"Also compute a severity-count summary across all findings (e.g. ",{"type":41,"tag":124,"props":1558,"children":1560},{"className":1559},[],[1561],{"type":47,"value":1562},"2 CRITICAL · 5 HIGH · 3 MEDIUM",{"type":47,"value":1564},") for the header of the report.",{"type":41,"tag":465,"props":1566,"children":1568},{"id":1567},"pulling-the-code-location",[1569],{"type":47,"value":1570},"Pulling the code location",{"type":41,"tag":50,"props":1572,"children":1573},{},[1574,1576,1582],{"type":47,"value":1575},"For each finding, derive a single short ",{"type":41,"tag":124,"props":1577,"children":1579},{"className":1578},[],[1580],{"type":47,"value":1581},"location",{"type":47,"value":1583}," string:",{"type":41,"tag":490,"props":1585,"children":1586},{},[1587,1600,1628],{"type":41,"tag":65,"props":1588,"children":1589},{},[1590,1592,1598],{"type":47,"value":1591},"If ",{"type":41,"tag":124,"props":1593,"children":1595},{"className":1594},[],[1596],{"type":47,"value":1597},"filePath",{"type":47,"value":1599}," is set, use it as-is.",{"type":41,"tag":65,"props":1601,"children":1602},{},[1603,1605,1611,1613,1618,1620,1626],{"type":47,"value":1604},"Otherwise, take ",{"type":41,"tag":124,"props":1606,"children":1608},{"className":1607},[],[1609],{"type":47,"value":1610},"codeLocations[0]",{"type":47,"value":1612},". Strip the scanner's sandbox prefix from ",{"type":41,"tag":124,"props":1614,"children":1616},{"className":1615},[],[1617],{"type":47,"value":1597},{"type":47,"value":1619},"\n(everything up to and including that marker) so the path is repo-relative; if that\nmarker isn't present, fall back to the basename. Append ",{"type":41,"tag":124,"props":1621,"children":1623},{"className":1622},[],[1624],{"type":47,"value":1625},":\u003ClineStart>",{"type":47,"value":1627}," when present.",{"type":41,"tag":65,"props":1629,"children":1630},{},[1631],{"type":47,"value":1632},"If neither is available (typical for some pentest findings), leave it blank and\ndescribe the affected endpoint or attack chain in the impact line instead.",{"type":41,"tag":465,"props":1634,"children":1636},{"id":1635},"summary-format",[1637],{"type":47,"value":1638},"Summary format",{"type":41,"tag":50,"props":1640,"children":1641},{},[1642],{"type":47,"value":1643},"Write a compact summary for the user:",{"type":41,"tag":157,"props":1645,"children":1648},{"className":1646,"code":1647,"language":47},[160],"## Security Agent triage — \u003Cagent space name>\n\n\u003CN> findings exported (\u003CP pentest, C code review>) · confidence: \u003Clevels> · severity: \u003Ccounts>\n\n### Priority order\n1. [CRITICAL · score 10.0 · HIGH confidence] \u003Cfinding name>\n   - Type: \u003CriskType> · Source: \u003Cpentest|code-review>\n   - Where: \u003Cfile:line or endpoint, if present>\n   - Impact: \u003Cone-line plain-language summary>\n2. [HIGH · ...] ...\n\n### Recommended remediation order\n\u003Cshort rationale: which to fix first and why — e.g. \"1 and 3 are both\nunauthenticated RCE on internet-facing endpoints; fix those before the\nstored-XSS issues.\">\n",[1649],{"type":41,"tag":124,"props":1650,"children":1651},{"__ignoreMap":165},[1652],{"type":47,"value":1647},{"type":41,"tag":50,"props":1654,"children":1655},{},[1656],{"type":47,"value":1657},"If more than ~10 findings, show the top N in detail and summarize the rest as a count\nby severity at the bottom.",{"type":41,"tag":465,"props":1659,"children":1661},{"id":1660},"what-to-keep-out-of-chat",[1662],{"type":47,"value":1663},"What to keep out of chat",{"type":41,"tag":50,"props":1665,"children":1666},{},[1667,1669,1674,1675,1681,1683,1689,1691,1696,1698,1704],{"type":47,"value":1668},"The full ",{"type":41,"tag":124,"props":1670,"children":1672},{"className":1671},[],[1673],{"type":47,"value":524},{"type":47,"value":427},{"type":41,"tag":124,"props":1676,"children":1678},{"className":1677},[],[1679],{"type":47,"value":1680},"reasoning",{"type":47,"value":1682},", and ",{"type":41,"tag":124,"props":1684,"children":1686},{"className":1685},[],[1687],{"type":47,"value":1688},"attackScript",{"type":47,"value":1690}," stay in the gitignored files —\nthey contain working exploit detail. In the chat summary keep impact lines to one line\neach, in plain language. Code-review findings usually carry a ",{"type":41,"tag":124,"props":1692,"children":1694},{"className":1693},[],[1695],{"type":47,"value":1597},{"type":47,"value":1697},"\u002Flocation and a\n",{"type":41,"tag":124,"props":1699,"children":1701},{"className":1700},[],[1702],{"type":47,"value":1703},"suggestedFix",{"type":47,"value":1705},"; call those out since they map directly to repo changes. Pentest findings\ndescribe endpoints and attack chains; map them to the responsible code where you can.\nLook for findings that corroborate each other (a pentest and a code review flagging the\nsame root cause) — those are strong signals for what to fix first.",{"type":41,"tag":107,"props":1707,"children":1709},{"id":1708},"stage-4-offer-to-remediate",[1710],{"type":47,"value":1711},"Stage 4: Offer to remediate",{"type":41,"tag":50,"props":1713,"children":1714},{},[1715],{"type":47,"value":1716},"After presenting the triage, offer to start fixing — don't silently begin editing code.",{"type":41,"tag":50,"props":1718,"children":1719},{},[1720,1722],{"type":47,"value":1721},"Ask the user something like: \"Want me to start fixing the top finding(s)? I'd recommend\nstarting with #1 (",{"type":41,"tag":516,"props":1723,"children":1724},{},[1725],{"type":47,"value":1726},").\" If they agree, work top-down by priority:",{"type":41,"tag":61,"props":1728,"children":1729},{},[1730,1735,1740],{"type":41,"tag":65,"props":1731,"children":1732},{},[1733],{"type":47,"value":1734},"Read the finding detail from the gitignored export file (location, description, suggested fix).",{"type":41,"tag":65,"props":1736,"children":1737},{},[1738],{"type":47,"value":1739},"Open the affected file and apply the fix via the editor.",{"type":41,"tag":65,"props":1741,"children":1742},{},[1743,1745,1751],{"type":47,"value":1744},"Report one line per fix: \"Fixed {name} in ",{"type":41,"tag":124,"props":1746,"children":1748},{"className":1747},[],[1749],{"type":47,"value":1750},"{filePath}:{lineStart}",{"type":47,"value":1752},".\"",{"type":41,"tag":50,"props":1754,"children":1755},{},[1756],{"type":47,"value":1757},"If the user wants to handle several findings, fix one at a time (or one cluster of related\nfindings) so each change stays reviewable, and proceed in the priority order from Stage 3.",{"type":41,"tag":107,"props":1759,"children":1761},{"id":1760},"notes-and-edge-cases",[1762],{"type":47,"value":1763},"Notes and edge cases",{"type":41,"tag":490,"props":1765,"children":1766},{},[1767,1784,1794,1804],{"type":41,"tag":65,"props":1768,"children":1769},{},[1770,1775,1777,1782],{"type":41,"tag":69,"props":1771,"children":1772},{},[1773],{"type":47,"value":1774},"No completed jobs",{"type":47,"value":1776},": a scan may still be ",{"type":41,"tag":124,"props":1778,"children":1780},{"className":1779},[],[1781],{"type":47,"value":425},{"type":47,"value":1783},". Tell the user; offer to re-check\nlater rather than exporting a partial job.",{"type":41,"tag":65,"props":1785,"children":1786},{},[1787,1792],{"type":41,"tag":69,"props":1788,"children":1789},{},[1790],{"type":47,"value":1791},"Re-running",{"type":47,"value":1793},": each run overwrites the files for that job id. The directory is safe to\ndelete; it only holds exported copies, not source-of-truth data.",{"type":41,"tag":65,"props":1795,"children":1796},{},[1797,1802],{"type":41,"tag":69,"props":1798,"children":1799},{},[1800],{"type":47,"value":1801},"Multiple accounts\u002FRegions",{"type":47,"value":1803},": findings are Region-scoped. If the user expected results\nand got none, confirm the region matches where Security Agent is configured.",{"type":41,"tag":65,"props":1805,"children":1806},{},[1807,1812],{"type":41,"tag":69,"props":1808,"children":1809},{},[1810],{"type":47,"value":1811},"Data handling",{"type":47,"value":1813},": treat exported findings as sensitive. They are copies of verified\nexploits against the user's own systems.",{"type":41,"tag":1815,"props":1816,"children":1817},"style",{},[1818],{"type":47,"value":1819},"html .light .shiki span {color: var(--shiki-light);background: var(--shiki-light-bg);font-style: var(--shiki-light-font-style);font-weight: var(--shiki-light-font-weight);text-decoration: var(--shiki-light-text-decoration);}html.light .shiki span {color: var(--shiki-light);background: var(--shiki-light-bg);font-style: var(--shiki-light-font-style);font-weight: var(--shiki-light-font-weight);text-decoration: var(--shiki-light-text-decoration);}html .default .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}html.dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}",{"items":1821,"total":1988},[1822,1839,1854,1867,1882,1892,1905,1921,1938,1951,1963,1978],{"slug":1823,"name":1823,"fn":1824,"description":1825,"org":1826,"tags":1827,"stars":24,"repoUrl":25,"updatedAt":1838},"agents-build","add capabilities to existing agent projects","Use when adding capabilities to an existing agent project — memory, app integration, VPC, multi-agent, migration, model changes, browser, code interpreter, or resource removal. Triggers on: \"add memory\", \"remember across sessions\", \"call agent from app\", \"invoke agent from code\", \"auth to call agent\", \"streaming responses\", \"VPC\", \"VPC connectivity\", \"VPC error\", \"can't reach from VPC\", \"multi-agent\", \"A2A\", \"A2A auth\", \"orchestrator not delegating\", \"specialist not called\", \"migrate Bedrock Agent\", \"after import\", \"migration issue\", \"framework for migration\", \"change model\", \"browser tool\", \"code interpreter\", \"delete agent\", \"tear down\", \"agentcore remove\", \"cross-account memory\", \"resource-based policy on memory\", \"pay for x402 content\", \"402 Payment Required\", \"microtransactions\", \"paid API or tool\". Not for connecting to external APIs via Gateway — use agents-connect. Not for scaffolding a new project — use agents-get-started. Not for CLI\u002Fdev server errors — use agents-debug. Strands vs LangGraph in a migration context routes here.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1828,1831,1834,1835],{"name":1829,"slug":1830,"type":15},"Agents","agents",{"name":1832,"slug":1833,"type":15},"Automation","automation",{"name":17,"slug":8,"type":15},{"name":1836,"slug":1837,"type":15},"Engineering","engineering","2026-07-12T08:42:53.812877",{"slug":1840,"name":1840,"fn":1841,"description":1842,"org":1843,"tags":1844,"stars":24,"repoUrl":25,"updatedAt":1853},"agents-connect","connect agents to external services","Use when connecting your agent to external APIs, tools, or services via Gateway, or restricting tool access with Cedar policies. Handles gateway setup, target types, outbound auth (OAuth, API key, IAM), credentials, and Cedar policy authoring. Triggers on: \"connect to API\", \"add gateway\", \"connect to MCP server\", \"Lambda tools\", \"OpenAPI\", \"gateway target\", \"Cedar policy\", \"restrict tools\", \"policy engine\", \"gateway auth error\", \"store API key\", \"outbound credential\", \"env var API key\", \"API key None after deploy\", \"credential not available after deploy\", \"should this be a gateway target\", \"give my agent tools\", \"add tools to agent\". Not for inbound auth (who can call your agent) — use agents-harden. Not for debugging agent behavior — use agents-debug. Not for VPC networking errors (agent can't reach APIs due to VPC) — use agents-build. Not for creating or hosting a new MCP server project — use agents-get-started.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1845,1846,1849,1852],{"name":1829,"slug":1830,"type":15},{"name":1847,"slug":1848,"type":15},"API Development","api-development",{"name":1850,"slug":1851,"type":15},"Authentication","authentication",{"name":17,"slug":8,"type":15},"2026-07-16T06:00:38.866147",{"slug":1855,"name":1855,"fn":1856,"description":1857,"org":1858,"tags":1859,"stars":24,"repoUrl":25,"updatedAt":1866},"agents-debug","debug agent and environment issues","Use when your agent or environment is broken — wrong answers, errors, timeouts, tool failures, or CLI issues. Reads traces and logs to diagnose root causes. Also checks prerequisites when the CLI itself isn't working. Triggers on: \"agent not working\", \"wrong answer\", \"agent error\", \"tool call failing\", \"debug agent\", \"check logs\", \"read traces\", \"broken\", \"500 error\", \"424 error\", \"model access denied\", \"command not found\", \"stuck in DELETING\", \"maxVms exceeded\", \"cold start diagnosis\", \"cold start slow\", \"agentcore create error\", \"create failed\", \"exit code 7\", \"connection refused local dev\". Not for deploy failures — use agents-deploy. Not for performance tuning without errors — use agents-optimize. Not for VPC configuration — use agents-build. Not for observability setup or missing logs — use agents-optimize.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1860,1861,1862,1863],{"name":1829,"slug":1830,"type":15},{"name":17,"slug":8,"type":15},{"name":22,"slug":23,"type":15},{"name":1864,"slug":1865,"type":15},"Observability","observability","2026-07-16T06:00:44.679093",{"slug":1868,"name":1868,"fn":1869,"description":1870,"org":1871,"tags":1872,"stars":24,"repoUrl":25,"updatedAt":1881},"agents-deploy","deploy AI agents to AWS","Use when deploying your agent to AWS, or when a deploy has failed. Handles pre-flight validation, CDK\u002FIAM\u002Fquota error diagnosis, version management, rollback, and canary deployments. Triggers on: \"deploy my agent\", \"agentcore deploy\", \"deploy failed\", \"CDK error\", \"rollback\", \"canary deploy\", \"pin version\", \"redeploy\", \"deploy stuck\". Not for production hardening — use agents-harden. Not for adding capabilities before deploy — use agents-build or agents-connect. Not for VPC configuration errors — use agents-build.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1873,1874,1875,1878],{"name":1829,"slug":1830,"type":15},{"name":17,"slug":8,"type":15},{"name":1876,"slug":1877,"type":15},"CI\u002FCD","ci-cd",{"name":1879,"slug":1880,"type":15},"Deployment","deployment","2026-07-12T08:42:55.059577",{"slug":1883,"name":1883,"fn":1884,"description":1885,"org":1886,"tags":1887,"stars":24,"repoUrl":25,"updatedAt":1891},"agents-get-started","scaffold and deploy new agent projects","Use when a developer wants to create a new agent project or get started with AgentCore. Handles framework selection, project scaffolding, first deploy, and first invocation. Triggers on: \"build an agent\", \"create an agent\", \"get started\", \"new project\", \"agentcore create\", \"which framework\", \"Strands vs LangGraph\", \"hello world agent\", \"first agent\", \"create MCP server\", \"host MCP server\", \"agentcore dev\", \"dev server\", \"what port\", \"local development\". Not for adding capabilities to existing projects — use agents-build or agents-connect. Strands vs LangGraph in a migration context routes to agents-build, not here. Connecting to an existing MCP server routes to agents-connect, not here.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1888,1889,1890],{"name":1829,"slug":1830,"type":15},{"name":17,"slug":8,"type":15},{"name":1879,"slug":1880,"type":15},"2026-07-12T08:42:51.963247",{"slug":1893,"name":1893,"fn":1894,"description":1895,"org":1896,"tags":1897,"stars":24,"repoUrl":25,"updatedAt":1904},"agents-harden","harden agents for production","Use when preparing your agent for production — IAM scoping, inbound auth (JWT, SigV4), secrets management, cold start optimization, session lifecycle, rate limiting, input validation, and quota guidance. Triggers on: \"production checklist\", \"harden agent\", \"production ready\", \"secure agent\", \"inbound auth\", \"going live\", \"cold start optimization\", \"session lifecycle\", \"StopRuntimeSession\", \"quota\", \"throttling\", \"maxVms\", \"rate limit\", \"security audit of outbound API calls\", \"gateway target audit for production\", \"restrict who can call\", \"lock down endpoint\", \"only our app can call\". Not for Cedar tool-restriction policies — use agents-connect. Not for quality measurement — use agents-optimize. Not for outbound credential storage or API key wiring — use agents-connect. Not for A2A agent-to-agent auth — use agents-build. Cold start observation and diagnosis (not optimization) routes to agents-debug.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1898,1899,1900,1903],{"name":1829,"slug":1830,"type":15},{"name":17,"slug":8,"type":15},{"name":1901,"slug":1902,"type":15},"Best Practices","best-practices",{"name":13,"slug":14,"type":15},"2026-07-16T06:00:42.174705",{"slug":1906,"name":1906,"fn":1907,"description":1908,"org":1909,"tags":1910,"stars":24,"repoUrl":25,"updatedAt":1920},"agents-optimize","optimize agent quality and performance","Use when measuring or improving agent quality and performance — set up evaluators, online monitoring, CI\u002FCD quality gates, observability, or cost optimization. Triggers on: \"evaluate my agent\", \"add evaluator\", \"measure quality\", \"quality gate\", \"run evals\", \"agent too slow\", \"why is it slow\", \"reduce latency\", \"set up observability\", \"CloudWatch dashboard\", \"how much does my agent cost\", \"cost optimization\", \"logs not showing up\", \"logs missing\", \"spans not found\", \"eval failing\", \"eval error\", \"dev traces\", \"local traces\", \"agentcore dev traces\", \"traces to CloudWatch\". Not for debugging errors or crashes — use agents-debug. Slow but correct routes here; broken routes to debug.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1911,1912,1913,1916,1917],{"name":1829,"slug":1830,"type":15},{"name":17,"slug":8,"type":15},{"name":1914,"slug":1915,"type":15},"Evals","evals",{"name":1864,"slug":1865,"type":15},{"name":1918,"slug":1919,"type":15},"Performance","performance","2026-07-12T08:42:56.488105",{"slug":1922,"name":1922,"fn":1923,"description":1924,"org":1925,"tags":1926,"stars":24,"repoUrl":25,"updatedAt":1937},"amazon-aurora-mysql","manage Amazon Aurora MySQL clusters","Amazon Aurora MySQL — creates, modifies, and advises on Aurora MySQL clusters specifically (MySQL-compatible engine, Aurora serverless, parallel query). Trigger for Aurora MySQL cluster operations, ACU sizing, I\u002FO-Optimized storage, commitment pricing, or MySQL upgrade planning. Aurora MySQL uses full (VPC-based) configuration — express configuration is PostgreSQL-only. For Aurora PostgreSQL, use amazon-aurora-postgresql instead. Contains safety guardrails and response templates that override defaults.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1927,1928,1931,1934],{"name":17,"slug":8,"type":15},{"name":1929,"slug":1930,"type":15},"Database","database",{"name":1932,"slug":1933,"type":15},"MySQL","mysql",{"name":1935,"slug":1936,"type":15},"Serverless","serverless","2026-07-12T08:43:13.27939",{"slug":1939,"name":1939,"fn":1940,"description":1941,"org":1942,"tags":1943,"stars":24,"repoUrl":25,"updatedAt":1950},"amazon-aurora-postgresql","configure Amazon Aurora PostgreSQL clusters","Amazon Aurora PostgreSQL — creates, modifies, and advises on Aurora PostgreSQL clusters specifically (PostgreSQL-compatible engine, Aurora serverless, express configuration, pgvector, Babelfish). Trigger for Aurora PostgreSQL cluster operations, express-configuration quick-start, ACU sizing, I\u002FO-Optimized storage, commitment pricing, or PostgreSQL upgrade planning. For Aurora MySQL, use amazon-aurora-mysql instead. Contains safety guardrails, express-first routing, and response templates that override defaults.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1944,1945,1946,1949],{"name":17,"slug":8,"type":15},{"name":1929,"slug":1930,"type":15},{"name":1947,"slug":1948,"type":15},"PostgreSQL","postgresql",{"name":1935,"slug":1936,"type":15},"2026-07-16T06:00:34.789624",{"slug":1952,"name":1952,"fn":1953,"description":1954,"org":1955,"tags":1956,"stars":24,"repoUrl":25,"updatedAt":1962},"amazon-bedrock","build generative AI apps with Amazon Bedrock","Builds generative AI applications on Amazon Bedrock. Covers model invocation (Converse API, InvokeModel), RAG with Knowledge Bases, Bedrock Agents, Guardrails, and AgentCore. Use when invoking models, setting up Knowledge Bases, creating agents, applying guardrails, deploying to AgentCore, migrating\u002Fporting\u002Fconverting a Bedrock Agent (including inline agents) to an AgentCore Harness, troubleshooting Bedrock errors (ThrottlingException, AccessDeniedException), or choosing models (Claude, Llama, Nova, Titan). ALSO USE for prompt caching setup and debugging, quota health checks and throttling diagnosis, cost attribution and tracking, migrating between Claude model generations (4.5 to 4.6 to 4.7), chunking strategies, API selection (Converse vs InvokeModel), guardrail capabilities, and model selection. Also covers AgentCore Payments setup (x402, microtransactions, Payment Manager, Connector, Instrument, Coinbase CDP, Stripe Privy, 402 Payment Required, pay for content, paid endpoint, agent payments). NOT for custom model training, Rekognition, or Comprehend.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1957,1958,1959],{"name":1829,"slug":1830,"type":15},{"name":17,"slug":8,"type":15},{"name":1960,"slug":1961,"type":15},"LLM","llm","2026-07-25T05:30:35.20899",{"slug":1964,"name":1964,"fn":1965,"description":1966,"org":1967,"tags":1968,"stars":24,"repoUrl":25,"updatedAt":1977},"amazon-documentdb","manage Amazon DocumentDB clusters","Manages Amazon DocumentDB end-to-end — serverless-on-8.0 cluster setup, TLS\u002FVPC\u002Fdriver config, flexible-schema and vector-search data modeling, MongoDB compatibility assessment, DMS-based migration, slow-query diagnosis, major version upgrades (4.0→5.0→8.0), Well-Architected reviews (41-check wa_review.py), cost estimation, and security hardening. Retrieve for every DocumentDB question and when the user asks to set up or migrate MongoDB to AWS — DocumentDB is AWS's MongoDB-compatible managed database. Triggers: JSON document store, document database, MongoDB on AWS, Nested fields, Lambda cannot connect, TLS handshake, VPC port 27017, IAM auth, Secrets Manager, encryption at rest, $graphLookup, flexible schema, COLLSCAN, compound index, DMS migration, CDC cutover, $vectorSearch, RAG, Global Clusters, DR replication, cost sizing, audit, health check, production-readiness.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1969,1970,1971,1974],{"name":17,"slug":8,"type":15},{"name":1929,"slug":1930,"type":15},{"name":1972,"slug":1973,"type":15},"MongoDB","mongodb",{"name":1975,"slug":1976,"type":15},"NoSQL","nosql","2026-07-12T08:43:00.455878",{"slug":1979,"name":1979,"fn":1980,"description":1981,"org":1982,"tags":1983,"stars":24,"repoUrl":25,"updatedAt":1987},"amazon-dynamodb","design and debug DynamoDB data layers","Designs, reviews, and debugs DynamoDB data layers from design axioms — enumerates access patterns, chooses partition\u002Fsort keys and GSIs, decides single-table vs. multi-table, configures Streams, Global Tables, TTL, and zero-ETL integrations to OpenSearch\u002FRedshift\u002FSageMaker Lakehouse, and produces a defensible data-layer design with a monthly cost estimate and optional live validation. Applies whenever a user is designing, reviewing, or refactoring anything backed by DynamoDB — schemas, access patterns, GSIs, single- vs. multi-table choices, Streams consumers, transactional outboxes, Global Tables, zero-ETL pipelines — even when they don't say \"axioms\" or \"design review.\" Also applies when debugging hot partitions, throttling, unbounded Scans, LWW conflicts, or surprise bills on DynamoDB workloads.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1984,1985,1986],{"name":17,"slug":8,"type":15},{"name":1929,"slug":1930,"type":15},{"name":1975,"slug":1976,"type":15},"2026-07-16T06:00:37.690386",115,{"items":1990,"total":2040},[1991,1998,2005,2012,2019,2025,2032],{"slug":1823,"name":1823,"fn":1824,"description":1825,"org":1992,"tags":1993,"stars":24,"repoUrl":25,"updatedAt":1838},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1994,1995,1996,1997],{"name":1829,"slug":1830,"type":15},{"name":1832,"slug":1833,"type":15},{"name":17,"slug":8,"type":15},{"name":1836,"slug":1837,"type":15},{"slug":1840,"name":1840,"fn":1841,"description":1842,"org":1999,"tags":2000,"stars":24,"repoUrl":25,"updatedAt":1853},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[2001,2002,2003,2004],{"name":1829,"slug":1830,"type":15},{"name":1847,"slug":1848,"type":15},{"name":1850,"slug":1851,"type":15},{"name":17,"slug":8,"type":15},{"slug":1855,"name":1855,"fn":1856,"description":1857,"org":2006,"tags":2007,"stars":24,"repoUrl":25,"updatedAt":1866},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[2008,2009,2010,2011],{"name":1829,"slug":1830,"type":15},{"name":17,"slug":8,"type":15},{"name":22,"slug":23,"type":15},{"name":1864,"slug":1865,"type":15},{"slug":1868,"name":1868,"fn":1869,"description":1870,"org":2013,"tags":2014,"stars":24,"repoUrl":25,"updatedAt":1881},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[2015,2016,2017,2018],{"name":1829,"slug":1830,"type":15},{"name":17,"slug":8,"type":15},{"name":1876,"slug":1877,"type":15},{"name":1879,"slug":1880,"type":15},{"slug":1883,"name":1883,"fn":1884,"description":1885,"org":2020,"tags":2021,"stars":24,"repoUrl":25,"updatedAt":1891},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[2022,2023,2024],{"name":1829,"slug":1830,"type":15},{"name":17,"slug":8,"type":15},{"name":1879,"slug":1880,"type":15},{"slug":1893,"name":1893,"fn":1894,"description":1895,"org":2026,"tags":2027,"stars":24,"repoUrl":25,"updatedAt":1904},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[2028,2029,2030,2031],{"name":1829,"slug":1830,"type":15},{"name":17,"slug":8,"type":15},{"name":1901,"slug":1902,"type":15},{"name":13,"slug":14,"type":15},{"slug":1906,"name":1906,"fn":1907,"description":1908,"org":2033,"tags":2034,"stars":24,"repoUrl":25,"updatedAt":1920},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[2035,2036,2037,2038,2039],{"name":1829,"slug":1830,"type":15},{"name":17,"slug":8,"type":15},{"name":1914,"slug":1915,"type":15},{"name":1864,"slug":1865,"type":15},{"name":1918,"slug":1919,"type":15},114]