[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-trail-of-bits-goal-prompt":3,"mdc--tj72gu-key":32,"related-org-trail-of-bits-goal-prompt":420,"related-repo-trail-of-bits-goal-prompt":582},{"slug":4,"name":4,"fn":5,"description":6,"org":7,"tags":12,"stars":20,"repoUrl":21,"updatedAt":22,"license":23,"forks":24,"topics":25,"repo":27,"sourceUrl":30,"mdContent":31},"goal-prompt","draft goal prompts for Claude Code","Drafts copy-paste-ready \u002Fgoal commands for goal mode in Claude Code and Codex. Use when the user asks to create, write, rewrite, improve, compress, clean up, or prepare a goal prompt, goal condition, \u002Fgoal command, goal-mode objective, or copy-ready long-running task objective.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},"trail-of-bits","Trail of Bits","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Ftrail-of-bits.png","trailofbits",[13,17],{"name":14,"slug":15,"type":16},"Prompt Engineering","prompt-engineering","tag",{"name":18,"slug":19,"type":16},"Claude Code","claude-code",6139,"https:\u002F\u002Fgithub.com\u002Ftrailofbits\u002Fskills","2026-08-21T03:59:00.541097",null,541,[26],"agent-skills",{"repoUrl":21,"stars":20,"forks":24,"topics":28,"description":29},[26],"Trail of Bits Claude Code skills for security research, vulnerability detection, and audit workflows","https:\u002F\u002Fgithub.com\u002Ftrailofbits\u002Fskills\u002Ftree\u002FHEAD\u002Fplugins\u002Fgoal-prompt\u002Fskills\u002Fgoal-prompt","---\nname: goal-prompt\ndescription: \"Drafts copy-paste-ready \u002Fgoal commands for goal mode in Claude Code and Codex. Use when the user asks to create, write, rewrite, improve, compress, clean up, or prepare a goal prompt, goal condition, \u002Fgoal command, goal-mode objective, or copy-ready long-running task objective.\"\nallowed-tools: Bash Read Write\n---\n\n# Goal Prompt\n\n`\u002Fgoal` keeps the agent working until a completion condition is met. Both Claude Code and Codex take it as one line, max 4,000 characters. In Claude Code a small model re-judges the condition after each turn from the transcript alone — it cannot run commands.\n\nDraft a condition that can terminate, then format it. A goal fits work bigger than one turn with a checkable finish line; chain small goals with review between them rather than writing one giant goal.\n\n## Draft\n\nInclude, joined with AND — never \"or\", the loop takes the cheaper branch:\n\n1. **End state, not activity** — \"all `legacyAuth()` call sites use `auth.verify()`\", not \"migrate the auth code\". An activity can be claimed; an end state is true or false.\n2. **Scope to read first** — the files, issue, logs, or plan to read before acting.\n3. **Stated check** — the exact command and its observable result (\"`npm test` exits 0\"), plus an instruction to run it and show the output; a result that never lands in the transcript does not exist to the evaluator.\n4. **Invariants** — what must not change (\"without modifying vendor\u002F\"), always including \"do not weaken, skip, or edit the checks themselves\".\n5. **Stop bound or blocked clause** — \"or stop after 20 turns\", \"if blocked, stop and report the blocker\". Without one, a mis-stated condition loops forever; the formatter warns when it is missing. (Claude Code resets the turn counter on session resume, so a turn bound silently extends across resumes.)\n\n**Keep it small.** Every constraint narrows the state space the model can explore. Collapse to one terminating criterion when possible, move scope and definitions into a referenced file, and drop non-goals — a constraint earns its place only by closing a real easy-out.\n\nFor long goals, also name the final evidence (diff, report, artifact) and require a progress log file — durable state across compaction and resume. If the brief exceeds 4,000 characters, put the details in a `GOAL.md` and reference that file from the objective.\n\n**Never invent missing elements.** Ground every element in the user's request, the conversation, or the repository — look things up rather than guessing. If an element cannot be filled from available information, still optimize and format what the user provided, leave the element out, and flag it as missing (see Format). A goal with an invented success condition terminates on the wrong contract.\n\n## Close the easy-outs\n\nBefore formatting, reread the drafted condition as a lazy model would: what is the cheapest way to make every check pass without doing the intended work? Close the cheapest ones — prefer pairing checks you already have over adding constraints, and do not enumerate every conceivable out into a non-goal list. The recurring outs:\n\n- **Delete or stub instead of fix** — \"search prints nothing\" also holds when the callers are gone; pair such checks with one that proves the feature still works.\n- **Pass on a subset** — running one test file, narrowing the search path, excluding directories from the check.\n- **Game the gate** — skipping\u002Fxfail-ing tests, hardcoding expected outputs, special-casing the test inputs, editing the check (the invariants rule).\n- **Claim without running** — declaring done or blocked with no check output in the transcript (the show-the-output rule).\n\nSame discipline as above: an out you cannot close from available information goes in the `Missing:` list as a warning — an invented or absurd constraint is worse than a flagged gap.\n\n## Security research goals\n\nCollapse audit goals to one terminating criterion, such as identifying, triggering, and validating one high-severity vulnerability valid under a referenced threat-model file. That file, not the goal, carries scope, attacker powers, severity baseline, and known findings to skip. Use neutral wording (\"trigger and validate\", not \"prove this is exploitable\"), require demonstrated preconditions — assumed attacker access is the most common false positive — and stop for human review after each finding rather than piling up untriaged reports. Validate findings with a second pass by a fresh agent, never the finder alone.\n\n## Format\n\nRun `uv run --no-project {baseDir}\u002Fscripts\u002Fformat_goal_prompt.py --fenced` on the draft (file or stdin). It collapses whitespace to one line, strips `\u002Fgoal` prefixes, quotes, and fences, warns on a missing stop clause, and rejects output over 4,000 characters — shorten or move detail to a file and rerun.\n\nReturn exactly one fenced `text` block, one line:\n\n```text\n\u002Fgoal \u003Csingle normalized objective>\n```\n\nAdd no prose around it — except when checklist elements could not be grounded: then follow the block with a `Missing:` list, one line per gap, telling the user what to supply.\n\n## Example\n\nDraft:\n\n```\n\u002Fgoal Migrate the auth module:\n  - replace legacyAuth() with auth.verify()\n  - make sure the tests still work\n```\n\nRedrafted and formatted:\n\n```text\n\u002Fgoal All legacyAuth() call sites use auth.verify(): `rg \"legacyAuth\\(\" -t ts` prints nothing AND `npm test` exits 0 (run both, show the output), without modifying vendor\u002F or weakening any test. If blocked, stop and report attempted paths and the blocker, or stop after 20 turns.\n```\n\nHere `npm test` came from the repo's package.json — not a guess — and pairing it with the zero-matches check closes the cheapest out: deleting the call sites instead of migrating them. When nothing grounds an element, format what exists and flag the gaps:\n\nDraft: `make checkout faster`, with no metric or benchmark anywhere in context:\n\n```text\n\u002Fgoal Make checkout faster\n```\n\nMissing:\n- measurable end state — which metric and threshold count as \"faster\"\n- verification — the benchmark or command that proves it\n- stop bound — e.g. \"or stop after 20 turns\"\n",{"data":33,"body":35},{"name":4,"description":6,"allowed-tools":34},"Bash Read Write",{"type":36,"children":37},"root",[38,46,59,64,71,76,156,166,179,189,195,200,244,257,263,268,274,294,306,318,330,336,341,350,355,364,376,389,398,402],{"type":39,"tag":40,"props":41,"children":42},"element","h1",{"id":4},[43],{"type":44,"value":45},"text","Goal Prompt",{"type":39,"tag":47,"props":48,"children":49},"p",{},[50,57],{"type":39,"tag":51,"props":52,"children":54},"code",{"className":53},[],[55],{"type":44,"value":56},"\u002Fgoal",{"type":44,"value":58}," keeps the agent working until a completion condition is met. Both Claude Code and Codex take it as one line, max 4,000 characters. In Claude Code a small model re-judges the condition after each turn from the transcript alone — it cannot run commands.",{"type":39,"tag":47,"props":60,"children":61},{},[62],{"type":44,"value":63},"Draft a condition that can terminate, then format it. A goal fits work bigger than one turn with a checkable finish line; chain small goals with review between them rather than writing one giant goal.",{"type":39,"tag":65,"props":66,"children":68},"h2",{"id":67},"draft",[69],{"type":44,"value":70},"Draft",{"type":39,"tag":47,"props":72,"children":73},{},[74],{"type":44,"value":75},"Include, joined with AND — never \"or\", the loop takes the cheaper branch:",{"type":39,"tag":77,"props":78,"children":79},"ol",{},[80,108,118,136,146],{"type":39,"tag":81,"props":82,"children":83},"li",{},[84,90,92,98,100,106],{"type":39,"tag":85,"props":86,"children":87},"strong",{},[88],{"type":44,"value":89},"End state, not activity",{"type":44,"value":91}," — \"all ",{"type":39,"tag":51,"props":93,"children":95},{"className":94},[],[96],{"type":44,"value":97},"legacyAuth()",{"type":44,"value":99}," call sites use ",{"type":39,"tag":51,"props":101,"children":103},{"className":102},[],[104],{"type":44,"value":105},"auth.verify()",{"type":44,"value":107},"\", not \"migrate the auth code\". An activity can be claimed; an end state is true or false.",{"type":39,"tag":81,"props":109,"children":110},{},[111,116],{"type":39,"tag":85,"props":112,"children":113},{},[114],{"type":44,"value":115},"Scope to read first",{"type":44,"value":117}," — the files, issue, logs, or plan to read before acting.",{"type":39,"tag":81,"props":119,"children":120},{},[121,126,128,134],{"type":39,"tag":85,"props":122,"children":123},{},[124],{"type":44,"value":125},"Stated check",{"type":44,"value":127}," — the exact command and its observable result (\"",{"type":39,"tag":51,"props":129,"children":131},{"className":130},[],[132],{"type":44,"value":133},"npm test",{"type":44,"value":135}," exits 0\"), plus an instruction to run it and show the output; a result that never lands in the transcript does not exist to the evaluator.",{"type":39,"tag":81,"props":137,"children":138},{},[139,144],{"type":39,"tag":85,"props":140,"children":141},{},[142],{"type":44,"value":143},"Invariants",{"type":44,"value":145}," — what must not change (\"without modifying vendor\u002F\"), always including \"do not weaken, skip, or edit the checks themselves\".",{"type":39,"tag":81,"props":147,"children":148},{},[149,154],{"type":39,"tag":85,"props":150,"children":151},{},[152],{"type":44,"value":153},"Stop bound or blocked clause",{"type":44,"value":155}," — \"or stop after 20 turns\", \"if blocked, stop and report the blocker\". Without one, a mis-stated condition loops forever; the formatter warns when it is missing. (Claude Code resets the turn counter on session resume, so a turn bound silently extends across resumes.)",{"type":39,"tag":47,"props":157,"children":158},{},[159,164],{"type":39,"tag":85,"props":160,"children":161},{},[162],{"type":44,"value":163},"Keep it small.",{"type":44,"value":165}," Every constraint narrows the state space the model can explore. Collapse to one terminating criterion when possible, move scope and definitions into a referenced file, and drop non-goals — a constraint earns its place only by closing a real easy-out.",{"type":39,"tag":47,"props":167,"children":168},{},[169,171,177],{"type":44,"value":170},"For long goals, also name the final evidence (diff, report, artifact) and require a progress log file — durable state across compaction and resume. If the brief exceeds 4,000 characters, put the details in a ",{"type":39,"tag":51,"props":172,"children":174},{"className":173},[],[175],{"type":44,"value":176},"GOAL.md",{"type":44,"value":178}," and reference that file from the objective.",{"type":39,"tag":47,"props":180,"children":181},{},[182,187],{"type":39,"tag":85,"props":183,"children":184},{},[185],{"type":44,"value":186},"Never invent missing elements.",{"type":44,"value":188}," Ground every element in the user's request, the conversation, or the repository — look things up rather than guessing. If an element cannot be filled from available information, still optimize and format what the user provided, leave the element out, and flag it as missing (see Format). A goal with an invented success condition terminates on the wrong contract.",{"type":39,"tag":65,"props":190,"children":192},{"id":191},"close-the-easy-outs",[193],{"type":44,"value":194},"Close the easy-outs",{"type":39,"tag":47,"props":196,"children":197},{},[198],{"type":44,"value":199},"Before formatting, reread the drafted condition as a lazy model would: what is the cheapest way to make every check pass without doing the intended work? Close the cheapest ones — prefer pairing checks you already have over adding constraints, and do not enumerate every conceivable out into a non-goal list. The recurring outs:",{"type":39,"tag":201,"props":202,"children":203},"ul",{},[204,214,224,234],{"type":39,"tag":81,"props":205,"children":206},{},[207,212],{"type":39,"tag":85,"props":208,"children":209},{},[210],{"type":44,"value":211},"Delete or stub instead of fix",{"type":44,"value":213}," — \"search prints nothing\" also holds when the callers are gone; pair such checks with one that proves the feature still works.",{"type":39,"tag":81,"props":215,"children":216},{},[217,222],{"type":39,"tag":85,"props":218,"children":219},{},[220],{"type":44,"value":221},"Pass on a subset",{"type":44,"value":223}," — running one test file, narrowing the search path, excluding directories from the check.",{"type":39,"tag":81,"props":225,"children":226},{},[227,232],{"type":39,"tag":85,"props":228,"children":229},{},[230],{"type":44,"value":231},"Game the gate",{"type":44,"value":233}," — skipping\u002Fxfail-ing tests, hardcoding expected outputs, special-casing the test inputs, editing the check (the invariants rule).",{"type":39,"tag":81,"props":235,"children":236},{},[237,242],{"type":39,"tag":85,"props":238,"children":239},{},[240],{"type":44,"value":241},"Claim without running",{"type":44,"value":243}," — declaring done or blocked with no check output in the transcript (the show-the-output rule).",{"type":39,"tag":47,"props":245,"children":246},{},[247,249,255],{"type":44,"value":248},"Same discipline as above: an out you cannot close from available information goes in the ",{"type":39,"tag":51,"props":250,"children":252},{"className":251},[],[253],{"type":44,"value":254},"Missing:",{"type":44,"value":256}," list as a warning — an invented or absurd constraint is worse than a flagged gap.",{"type":39,"tag":65,"props":258,"children":260},{"id":259},"security-research-goals",[261],{"type":44,"value":262},"Security research goals",{"type":39,"tag":47,"props":264,"children":265},{},[266],{"type":44,"value":267},"Collapse audit goals to one terminating criterion, such as identifying, triggering, and validating one high-severity vulnerability valid under a referenced threat-model file. That file, not the goal, carries scope, attacker powers, severity baseline, and known findings to skip. Use neutral wording (\"trigger and validate\", not \"prove this is exploitable\"), require demonstrated preconditions — assumed attacker access is the most common false positive — and stop for human review after each finding rather than piling up untriaged reports. Validate findings with a second pass by a fresh agent, never the finder alone.",{"type":39,"tag":65,"props":269,"children":271},{"id":270},"format",[272],{"type":44,"value":273},"Format",{"type":39,"tag":47,"props":275,"children":276},{},[277,279,285,287,292],{"type":44,"value":278},"Run ",{"type":39,"tag":51,"props":280,"children":282},{"className":281},[],[283],{"type":44,"value":284},"uv run --no-project {baseDir}\u002Fscripts\u002Fformat_goal_prompt.py --fenced",{"type":44,"value":286}," on the draft (file or stdin). It collapses whitespace to one line, strips ",{"type":39,"tag":51,"props":288,"children":290},{"className":289},[],[291],{"type":44,"value":56},{"type":44,"value":293}," prefixes, quotes, and fences, warns on a missing stop clause, and rejects output over 4,000 characters — shorten or move detail to a file and rerun.",{"type":39,"tag":47,"props":295,"children":296},{},[297,299,304],{"type":44,"value":298},"Return exactly one fenced ",{"type":39,"tag":51,"props":300,"children":302},{"className":301},[],[303],{"type":44,"value":44},{"type":44,"value":305}," block, one line:",{"type":39,"tag":307,"props":308,"children":313},"pre",{"className":309,"code":311,"language":44,"meta":312},[310],"language-text","\u002Fgoal \u003Csingle normalized objective>\n","",[314],{"type":39,"tag":51,"props":315,"children":316},{"__ignoreMap":312},[317],{"type":44,"value":311},{"type":39,"tag":47,"props":319,"children":320},{},[321,323,328],{"type":44,"value":322},"Add no prose around it — except when checklist elements could not be grounded: then follow the block with a ",{"type":39,"tag":51,"props":324,"children":326},{"className":325},[],[327],{"type":44,"value":254},{"type":44,"value":329}," list, one line per gap, telling the user what to supply.",{"type":39,"tag":65,"props":331,"children":333},{"id":332},"example",[334],{"type":44,"value":335},"Example",{"type":39,"tag":47,"props":337,"children":338},{},[339],{"type":44,"value":340},"Draft:",{"type":39,"tag":307,"props":342,"children":345},{"className":343,"code":344,"language":44},[310],"\u002Fgoal Migrate the auth module:\n  - replace legacyAuth() with auth.verify()\n  - make sure the tests still work\n",[346],{"type":39,"tag":51,"props":347,"children":348},{"__ignoreMap":312},[349],{"type":44,"value":344},{"type":39,"tag":47,"props":351,"children":352},{},[353],{"type":44,"value":354},"Redrafted and formatted:",{"type":39,"tag":307,"props":356,"children":359},{"className":357,"code":358,"language":44,"meta":312},[310],"\u002Fgoal All legacyAuth() call sites use auth.verify(): `rg \"legacyAuth\\(\" -t ts` prints nothing AND `npm test` exits 0 (run both, show the output), without modifying vendor\u002F or weakening any test. If blocked, stop and report attempted paths and the blocker, or stop after 20 turns.\n",[360],{"type":39,"tag":51,"props":361,"children":362},{"__ignoreMap":312},[363],{"type":44,"value":358},{"type":39,"tag":47,"props":365,"children":366},{},[367,369,374],{"type":44,"value":368},"Here ",{"type":39,"tag":51,"props":370,"children":372},{"className":371},[],[373],{"type":44,"value":133},{"type":44,"value":375}," came from the repo's package.json — not a guess — and pairing it with the zero-matches check closes the cheapest out: deleting the call sites instead of migrating them. When nothing grounds an element, format what exists and flag the gaps:",{"type":39,"tag":47,"props":377,"children":378},{},[379,381,387],{"type":44,"value":380},"Draft: ",{"type":39,"tag":51,"props":382,"children":384},{"className":383},[],[385],{"type":44,"value":386},"make checkout faster",{"type":44,"value":388},", with no metric or benchmark anywhere in context:",{"type":39,"tag":307,"props":390,"children":393},{"className":391,"code":392,"language":44,"meta":312},[310],"\u002Fgoal Make checkout faster\n",[394],{"type":39,"tag":51,"props":395,"children":396},{"__ignoreMap":312},[397],{"type":44,"value":392},{"type":39,"tag":47,"props":399,"children":400},{},[401],{"type":44,"value":254},{"type":39,"tag":201,"props":403,"children":404},{},[405,410,415],{"type":39,"tag":81,"props":406,"children":407},{},[408],{"type":44,"value":409},"measurable end state — which metric and threshold count as \"faster\"",{"type":39,"tag":81,"props":411,"children":412},{},[413],{"type":44,"value":414},"verification — the benchmark or command that proves it",{"type":39,"tag":81,"props":416,"children":417},{},[418],{"type":44,"value":419},"stop bound — e.g. \"or stop after 20 turns\"",{"items":421,"total":581},[422,441,451,471,486,498,508,523,534,546,557,568],{"slug":423,"name":423,"fn":424,"description":425,"org":426,"tags":427,"stars":20,"repoUrl":21,"updatedAt":440},"address-sanitizer","detect memory errors during fuzzing","AddressSanitizer detects memory errors during fuzzing. Use when fuzzing C\u002FC++ code to find buffer overflows and use-after-free bugs.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[428,431,434,437],{"name":429,"slug":430,"type":16},"C#","c",{"name":432,"slug":433,"type":16},"Debugging","debugging",{"name":435,"slug":436,"type":16},"Security","security",{"name":438,"slug":439,"type":16},"Testing","testing","2026-07-17T06:05:14.925095",{"slug":442,"name":442,"fn":443,"description":444,"org":445,"tags":446,"stars":20,"repoUrl":21,"updatedAt":450},"aflpp","perform multi-core fuzzing of C\u002FC++ projects","AFL++ is a fork of AFL with better fuzzing performance and advanced features. Use for multi-core fuzzing of C\u002FC++ projects.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[447,448,449],{"name":429,"slug":430,"type":16},{"name":435,"slug":436,"type":16},{"name":438,"slug":439,"type":16},"2026-07-17T06:05:12.433192",{"slug":452,"name":452,"fn":453,"description":454,"org":455,"tags":456,"stars":20,"repoUrl":21,"updatedAt":470},"agentic-actions-auditor","audit GitHub Actions for security vulnerabilities","Audits GitHub Actions workflows for security vulnerabilities in AI agent integrations including Claude Code Action, Gemini CLI, OpenAI Codex, and GitHub AI Inference. Detects attack vectors where attacker-controlled input reaches AI agents running in CI\u002FCD pipelines, including env var intermediary patterns, direct expression injection, dangerous sandbox configurations, and wildcard user allowlists. Use when reviewing workflow files that invoke AI coding agents, auditing CI\u002FCD pipeline security for prompt injection risks, or evaluating agentic action configurations.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[457,460,463,466,469],{"name":458,"slug":459,"type":16},"Agents","agents",{"name":461,"slug":462,"type":16},"CI\u002FCD","ci-cd",{"name":464,"slug":465,"type":16},"Code Analysis","code-analysis",{"name":467,"slug":468,"type":16},"GitHub Actions","github-actions",{"name":435,"slug":436,"type":16},"2026-07-18T05:47:48.564744",{"slug":472,"name":472,"fn":473,"description":474,"org":475,"tags":476,"stars":20,"repoUrl":21,"updatedAt":485},"algorand-vulnerability-scanner","scan Algorand smart contracts for vulnerabilities","Scans Algorand smart contracts for 11 common vulnerabilities including rekeying attacks, unchecked transaction fees, missing field validations, and access control issues. Use when auditing Algorand projects (TEAL\u002FPyTeal).",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[477,480,481,482],{"name":478,"slug":479,"type":16},"Audit","audit",{"name":464,"slug":465,"type":16},{"name":435,"slug":436,"type":16},{"name":483,"slug":484,"type":16},"Smart Contracts","smart-contracts","2026-08-21T03:32:34.227368",{"slug":487,"name":487,"fn":488,"description":489,"org":490,"tags":491,"stars":20,"repoUrl":21,"updatedAt":497},"atheris","fuzz Python code with Atheris","Atheris is a coverage-guided Python fuzzer based on libFuzzer. Use for fuzzing pure Python code and Python C extensions.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[492,495,496],{"name":493,"slug":494,"type":16},"Python","python",{"name":435,"slug":436,"type":16},{"name":438,"slug":439,"type":16},"2026-08-21T03:32:31.260317",{"slug":499,"name":499,"fn":500,"description":501,"org":502,"tags":503,"stars":20,"repoUrl":21,"updatedAt":507},"audit-augmentation","augment code graphs with audit findings","Augments Trailmark code graphs with external audit findings from SARIF static analysis results, weAudit annotation files, and version-gated Trailmark 0.4.x binary-analysis graph exports. Maps findings to graph nodes by file and line overlap, creates severity-based subgraphs, and enables cross-referencing findings with pre-analysis data (blast radius, taint, etc.). Use when projecting SARIF results onto a code graph, overlaying weAudit annotations, importing binary graph findings, cross-referencing Semgrep, CodeQL, or binary-analysis findings with call graph data, or visualizing audit findings in the context of code structure.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[504,505,506],{"name":478,"slug":479,"type":16},{"name":464,"slug":465,"type":16},{"name":435,"slug":436,"type":16},"2026-08-21T03:32:38.432332",{"slug":509,"name":509,"fn":510,"description":511,"org":512,"tags":513,"stars":20,"repoUrl":21,"updatedAt":522},"audit-context-building","build architectural context for code analysis","Understand a codebase before looking for bugs in it - what each function assumes, what it guarantees, and what it depends on elsewhere. Use when starting an audit, threat model, or architecture review on unfamiliar code, and before any vulnerability-hunting pass.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[514,517,518,519],{"name":515,"slug":516,"type":16},"Architecture","architecture",{"name":478,"slug":479,"type":16},{"name":464,"slug":465,"type":16},{"name":520,"slug":521,"type":16},"Engineering","engineering","2026-08-08T03:55:04.309014",{"slug":524,"name":524,"fn":525,"description":526,"org":527,"tags":528,"stars":20,"repoUrl":21,"updatedAt":533},"audit-prep-assistant","prepare codebases for security audits","Prepares codebases for security review using Trail of Bits' checklist. Helps set review goals, runs static analysis tools, increases test coverage, removes dead code, ensures accessibility, and generates documentation (flowcharts, user stories, inline comments).",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[529,530,531,532],{"name":478,"slug":479,"type":16},{"name":464,"slug":465,"type":16},{"name":520,"slug":521,"type":16},{"name":435,"slug":436,"type":16},"2026-07-18T05:47:39.210985",{"slug":535,"name":535,"fn":536,"description":537,"org":538,"tags":539,"stars":20,"repoUrl":21,"updatedAt":545},"burpsuite-project-parser","parse Burp Suite project files","Searches and explores Burp Suite project files (.burp) from the command line. Use when searching response headers or bodies with regex patterns, extracting security audit findings, dumping proxy history or site map data, or analyzing HTTP traffic captured in a Burp project.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[540,541,544],{"name":478,"slug":479,"type":16},{"name":542,"slug":543,"type":16},"CLI","cli",{"name":435,"slug":436,"type":16},"2026-07-17T06:05:33.198077",{"slug":547,"name":547,"fn":548,"description":549,"org":550,"tags":551,"stars":20,"repoUrl":21,"updatedAt":556},"c-review","audit C and C++ code","Performs comprehensive C\u002FC++ security review for memory corruption, integer overflows, race conditions, and platform-specific vulnerabilities. Use when auditing native C\u002FC++ applications, reviewing daemons or services for memory safety, or hunting integer overflow \u002F use-after-free \u002F race conditions in userspace code.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[552,553,554,555],{"name":478,"slug":479,"type":16},{"name":429,"slug":430,"type":16},{"name":464,"slug":465,"type":16},{"name":435,"slug":436,"type":16},"2026-08-21T03:32:30.26169",{"slug":558,"name":558,"fn":559,"description":560,"org":561,"tags":562,"stars":20,"repoUrl":21,"updatedAt":567},"cairo-vulnerability-scanner","scan Cairo and StarkNet contracts for vulnerabilities","Scans Cairo\u002FStarkNet smart contracts for 6 critical vulnerabilities including felt252 arithmetic overflow, L1-L2 messaging issues, address conversion problems, and signature replay. Use when auditing StarkNet projects.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[563,564,565,566],{"name":478,"slug":479,"type":16},{"name":464,"slug":465,"type":16},{"name":435,"slug":436,"type":16},{"name":483,"slug":484,"type":16},"2026-08-21T03:32:39.232318",{"slug":569,"name":569,"fn":570,"description":571,"org":572,"tags":573,"stars":20,"repoUrl":21,"updatedAt":580},"cargo-fuzz","fuzz Rust projects with cargo-fuzz","cargo-fuzz is the de facto fuzzing tool for Rust projects using Cargo. Use for fuzzing Rust code with libFuzzer backend.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[574,575,578,579],{"name":520,"slug":521,"type":16},{"name":576,"slug":577,"type":16},"Rust","rust",{"name":435,"slug":436,"type":16},{"name":438,"slug":439,"type":16},"2026-07-17T06:05:18.942184",110,{"items":583,"total":631},[584,591,597,605,612,618,624],{"slug":423,"name":423,"fn":424,"description":425,"org":585,"tags":586,"stars":20,"repoUrl":21,"updatedAt":440},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[587,588,589,590],{"name":429,"slug":430,"type":16},{"name":432,"slug":433,"type":16},{"name":435,"slug":436,"type":16},{"name":438,"slug":439,"type":16},{"slug":442,"name":442,"fn":443,"description":444,"org":592,"tags":593,"stars":20,"repoUrl":21,"updatedAt":450},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[594,595,596],{"name":429,"slug":430,"type":16},{"name":435,"slug":436,"type":16},{"name":438,"slug":439,"type":16},{"slug":452,"name":452,"fn":453,"description":454,"org":598,"tags":599,"stars":20,"repoUrl":21,"updatedAt":470},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[600,601,602,603,604],{"name":458,"slug":459,"type":16},{"name":461,"slug":462,"type":16},{"name":464,"slug":465,"type":16},{"name":467,"slug":468,"type":16},{"name":435,"slug":436,"type":16},{"slug":472,"name":472,"fn":473,"description":474,"org":606,"tags":607,"stars":20,"repoUrl":21,"updatedAt":485},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[608,609,610,611],{"name":478,"slug":479,"type":16},{"name":464,"slug":465,"type":16},{"name":435,"slug":436,"type":16},{"name":483,"slug":484,"type":16},{"slug":487,"name":487,"fn":488,"description":489,"org":613,"tags":614,"stars":20,"repoUrl":21,"updatedAt":497},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[615,616,617],{"name":493,"slug":494,"type":16},{"name":435,"slug":436,"type":16},{"name":438,"slug":439,"type":16},{"slug":499,"name":499,"fn":500,"description":501,"org":619,"tags":620,"stars":20,"repoUrl":21,"updatedAt":507},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[621,622,623],{"name":478,"slug":479,"type":16},{"name":464,"slug":465,"type":16},{"name":435,"slug":436,"type":16},{"slug":509,"name":509,"fn":510,"description":511,"org":625,"tags":626,"stars":20,"repoUrl":21,"updatedAt":522},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[627,628,629,630],{"name":515,"slug":516,"type":16},{"name":478,"slug":479,"type":16},{"name":464,"slug":465,"type":16},{"name":520,"slug":521,"type":16},76]