[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-pydantic-pydantic-ai-harness":3,"mdc--x3ohww-key":34,"related-repo-pydantic-pydantic-ai-harness":1271,"related-org-pydantic-pydantic-ai-harness":1348},{"slug":4,"name":4,"fn":5,"description":6,"org":7,"tags":11,"stars":23,"repoUrl":24,"updatedAt":25,"license":26,"forks":27,"topics":28,"repo":29,"sourceUrl":32,"mdContent":33},"pydantic-ai-harness","extend Pydantic AI agents with capabilities","Extend Pydantic AI agents with batteries-included capabilities from pydantic-ai-harness -- Code Mode (collapse many tool calls into one sandboxed Python execution), a filesystem and shell, sub-agents, planning, context compaction, and more. Use when the user mentions pydantic-ai-harness, CodeMode, Monty, code mode, or tool sandboxing, when they want first-party filesystem\u002Fshell\u002Fsub-agent\u002Fplanning\u002Fcompaction capabilities for a Pydantic AI agent, when they want an agent to run agent-written Python, or when a Pydantic AI agent would benefit from orchestrating multiple tool calls in a single sandboxed script.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},"pydantic","Pydantic","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Fpydantic.png",[12,14,17,20],{"name":9,"slug":8,"type":13},"tag",{"name":15,"slug":16,"type":13},"Python","python",{"name":18,"slug":19,"type":13},"Agents","agents",{"name":21,"slug":22,"type":13},"Code Execution","code-execution",95,"https:\u002F\u002Fgithub.com\u002Fpydantic\u002Fskills","2026-07-30T05:28:54.443013","MIT",3,[],{"repoUrl":24,"stars":23,"forks":27,"topics":30,"description":31},[],null,"https:\u002F\u002Fgithub.com\u002Fpydantic\u002Fskills\u002Ftree\u002FHEAD\u002Fskills\u002Fpydantic-ai-harness","---\nname: pydantic-ai-harness\ndescription: Extend Pydantic AI agents with batteries-included capabilities from pydantic-ai-harness -- Code Mode (collapse many tool calls into one sandboxed Python execution), a filesystem and shell, sub-agents, planning, context compaction, and more. Use when the user mentions pydantic-ai-harness, CodeMode, Monty, code mode, or tool sandboxing, when they want first-party filesystem\u002Fshell\u002Fsub-agent\u002Fplanning\u002Fcompaction capabilities for a Pydantic AI agent, when they want an agent to run agent-written Python, or when a Pydantic AI agent would benefit from orchestrating multiple tool calls in a single sandboxed script.\nlicense: MIT\ncompatibility: Requires Python 3.10+ and pydantic-ai-slim>=2.18.0\nmetadata:\n  version: \"0.1.0\"\n  author: pydantic\n---\n\n# Building with Pydantic AI Harness\n\nPydantic AI Harness is the official capability library for Pydantic AI. Capabilities that need model or\nframework support -- and those fundamental to every agent -- live in core `pydantic-ai`; optional,\nbatteries-included capabilities live here. Both are composed onto an agent through the same\n`capabilities=[...]` API.\n\nThis skill covers the capabilities shipped by `pydantic-ai-harness`. For the core framework -- agents,\ntools, structured output, hooks, and testing -- use the `building-pydantic-ai-agents` skill instead.\n\n## When to Use This Skill\n\nInvoke this skill when:\n- The user mentions `pydantic-ai-harness`, `CodeMode`, code mode, or the Monty sandbox\n- An agent makes many sequential tool calls that could collapse into one sandboxed Python execution\n- The user wants the model to write Python that loops, branches, aggregates, or parallelizes tool calls with `asyncio.gather`\n- The user asks to sandbox or constrain the code an agent runs\n\nDo **not** use this skill for:\n- Core Pydantic AI usage -- building agents, adding tools, structured output, streaming, or testing (use `building-pydantic-ai-agents`)\n- Capabilities that ship in core `pydantic-ai`, such as web search, tool search, and thinking\n- The Pydantic validation library on its own (`pydantic`\u002F`BaseModel` without agents)\n\n## Supported Capabilities\n\n`CodeMode` has a full reference below; it is the flagship capability and the one this skill goes deep on.\nThe rest ship today and each has its own README with API and examples.\n\nEach capability lives in its own submodule and is imported from there\n(`from pydantic_ai_harness.\u003Cmodule> import ...`). Capabilities are not importable from the top-level\n`pydantic_ai_harness` package by design, so each one keeps its own optional dependencies isolated.\n`CodeMode`, `FileSystem`, `Shell`, and `ManagedPrompt` also have top-level re-exports (importable directly\nfrom `pydantic_ai_harness`).\n\nAPIs are subject to change between releases; breaking changes ship deprecation warnings where practical.\n\n| Capability | Module | Description |\n|---|---|---|\n| `CodeMode` | `pydantic_ai_harness.code_mode` (also top-level) | Wraps eligible tools into a single sandboxed `run_code` tool so the model orchestrates them in Python -- see [Code Mode](.\u002Freferences\u002FCODE-MODE.md) |\n| `FileSystem` | `pydantic_ai_harness.filesystem` (also top-level) | Read, write, edit, and search files under a root directory, with traversal prevention |\n| `Shell` | `pydantic_ai_harness.shell` (also top-level) | Run commands in a subprocess with allowlists, a default denylist, timeouts, and env masking |\n| `ManagedPrompt` | `pydantic_ai_harness.logfire` (also top-level) | Back an agent's instructions with a Logfire-managed prompt |\n| `SubAgents` | `pydantic_ai_harness.subagents` | Delegate subtasks to specialized child agents |\n| `DynamicWorkflow` | `pydantic_ai_harness.dynamic_workflow` | Orchestrate sub-agents from a model-written Python script |\n| `Planning` | `pydantic_ai_harness.planning` | Break complex tasks into structured plans before execution |\n| compaction family (`SlidingWindowCompaction`, `SummarizingCompaction`, ...) | `pydantic_ai_harness.compaction` | Trim or summarize conversation history to stay within token limits |\n| `ToolOutputLimits` | `pydantic_ai_harness.tool_output_limits` | Truncate, summarize, or spill large tool outputs |\n| `RepoContext` | `pydantic_ai_harness.repo_context` | Auto-load CLAUDE.md\u002FAGENTS.md and repo structure |\n| `StepPersistence` | `pydantic_ai_harness.step_persistence` | Save, restore, resume, and fork run state |\n| `PydanticAIDocs` | `pydantic_ai_harness.pydantic_ai_docs` | On-demand `read_pyai_docs` tool for Pydantic AI docs |\n| `CapabilityCreation` | `pydantic_ai_harness.capability_creation` | Let an agent author, validate, and load real capabilities at runtime |\n| media externalization | `pydantic_ai_harness.media` | Offload large `BinaryContent` to content-addressed stores |\n\nStill experimental: an ACP server adapter, imported from `pydantic_ai_harness.experimental.acp`. Importing it\nemits a `HarnessExperimentalWarning`.\n\nThe full, current list with links and status is in the\n[capability matrix](https:\u002F\u002Fgithub.com\u002Fpydantic\u002Fpydantic-ai-harness#capability-matrix).\n\n## Install\n\n```bash\nuv add pydantic-ai-harness\n```\n\nEach capability declares its own extra. Code Mode needs the Monty sandbox:\n\n```bash\nuv add \"pydantic-ai-harness[codemode]\"   # `code-mode` is also accepted as an alias\n```\n\nRequires Python 3.10+ and `pydantic-ai-slim>=2.18.0`.\n\n## Quick Start\n\nA harness capability is added to the agent like any other. Here `CodeMode` wraps locally registered tools\ninto a single `run_code` tool that the model drives with Python.\n\n```python {test=\"skip\"}\nfrom pydantic_ai import Agent\n\nfrom pydantic_ai_harness import CodeMode\n\nagent = Agent('anthropic:claude-sonnet-4-6', capabilities=[CodeMode()])\n\n\n@agent.tool_plain\ndef get_temperature_f(city: str) -> float:\n    return {'Paris': 68.0, 'Tokyo': 77.0}[city]\n\n\n@agent.tool_plain\ndef convert_temp(fahrenheit: float) -> float:\n    return round((fahrenheit - 32) * 5 \u002F 9, 1)\n\nresult = agent.run_sync(\n    'Compare the weather in Paris and Tokyo, and report both temperatures in Celsius.'\n)\nprint(result.output)\n#> Paris is 20.0 C and Tokyo is 25.0 C.\n```\n\nThe model writes a single Python script that fetches both temperatures with `asyncio.gather` and then\nconverts them -- performing four tool calls across two dependent stages in one `run_code` invocation.\n\n## Key Practices\n\n- **Confirm a harness capability is actually needed.** If core Pydantic AI tools and capabilities are enough, use the `building-pydantic-ai-agents` skill instead -- don't reach for the harness by default.\n- **Read the reference before writing code.** Each capability has its own configuration, constraints, and gotchas -- load the linked reference (e.g. [Code Mode](.\u002Freferences\u002FCODE-MODE.md)) first.\n- **Install the capability's extra.** Importing `CodeMode` without `pydantic-ai-harness[codemode]` raises an `ImportError`; the Monty sandbox is an optional dependency.\n\n## Common Gotchas\n\n- **`native=True` tools bypass `CodeMode`.** Provider-native MCP servers and web search execute server-side, so `run_code` never sees them. Use `native=False` for client-side dispatch that `CodeMode` can wrap, but do not treat a remote server as trusted or sandboxed; see the [Code Mode trust boundary](.\u002Freferences\u002FCODE-MODE.md#sandbox-restrictions).\n- **The Monty sandbox is a Python subset.** It has no third-party imports and only a small stdlib allowlist -- read [Code Mode](.\u002Freferences\u002FCODE-MODE.md#sandbox-restrictions) before debugging generated code that fails to run.\n- **`CodeMode` needs its extra.** Install `pydantic-ai-harness[codemode]`, not the bare package.\n",{"data":35,"body":39},{"name":4,"description":6,"license":26,"compatibility":36,"metadata":37},"Requires Python 3.10+ and pydantic-ai-slim>=2.18.0",{"version":38,"author":8},"0.1.0",{"type":40,"children":41},"root",[42,51,74,94,101,106,152,165,212,218,228,285,290,726,747,761,767,801,806,846,858,864,883,1074,1093,1099,1167,1173,1265],{"type":43,"tag":44,"props":45,"children":47},"element","h1",{"id":46},"building-with-pydantic-ai-harness",[48],{"type":49,"value":50},"text","Building with Pydantic AI Harness",{"type":43,"tag":52,"props":53,"children":54},"p",{},[55,57,64,66,72],{"type":49,"value":56},"Pydantic AI Harness is the official capability library for Pydantic AI. Capabilities that need model or\nframework support -- and those fundamental to every agent -- live in core ",{"type":43,"tag":58,"props":59,"children":61},"code",{"className":60},[],[62],{"type":49,"value":63},"pydantic-ai",{"type":49,"value":65},"; optional,\nbatteries-included capabilities live here. Both are composed onto an agent through the same\n",{"type":43,"tag":58,"props":67,"children":69},{"className":68},[],[70],{"type":49,"value":71},"capabilities=[...]",{"type":49,"value":73}," API.",{"type":43,"tag":52,"props":75,"children":76},{},[77,79,84,86,92],{"type":49,"value":78},"This skill covers the capabilities shipped by ",{"type":43,"tag":58,"props":80,"children":82},{"className":81},[],[83],{"type":49,"value":4},{"type":49,"value":85},". For the core framework -- agents,\ntools, structured output, hooks, and testing -- use the ",{"type":43,"tag":58,"props":87,"children":89},{"className":88},[],[90],{"type":49,"value":91},"building-pydantic-ai-agents",{"type":49,"value":93}," skill instead.",{"type":43,"tag":95,"props":96,"children":98},"h2",{"id":97},"when-to-use-this-skill",[99],{"type":49,"value":100},"When to Use This Skill",{"type":43,"tag":52,"props":102,"children":103},{},[104],{"type":49,"value":105},"Invoke this skill when:",{"type":43,"tag":107,"props":108,"children":109},"ul",{},[110,131,136,147],{"type":43,"tag":111,"props":112,"children":113},"li",{},[114,116,121,123,129],{"type":49,"value":115},"The user mentions ",{"type":43,"tag":58,"props":117,"children":119},{"className":118},[],[120],{"type":49,"value":4},{"type":49,"value":122},", ",{"type":43,"tag":58,"props":124,"children":126},{"className":125},[],[127],{"type":49,"value":128},"CodeMode",{"type":49,"value":130},", code mode, or the Monty sandbox",{"type":43,"tag":111,"props":132,"children":133},{},[134],{"type":49,"value":135},"An agent makes many sequential tool calls that could collapse into one sandboxed Python execution",{"type":43,"tag":111,"props":137,"children":138},{},[139,141],{"type":49,"value":140},"The user wants the model to write Python that loops, branches, aggregates, or parallelizes tool calls with ",{"type":43,"tag":58,"props":142,"children":144},{"className":143},[],[145],{"type":49,"value":146},"asyncio.gather",{"type":43,"tag":111,"props":148,"children":149},{},[150],{"type":49,"value":151},"The user asks to sandbox or constrain the code an agent runs",{"type":43,"tag":52,"props":153,"children":154},{},[155,157,163],{"type":49,"value":156},"Do ",{"type":43,"tag":158,"props":159,"children":160},"strong",{},[161],{"type":49,"value":162},"not",{"type":49,"value":164}," use this skill for:",{"type":43,"tag":107,"props":166,"children":167},{},[168,180,192],{"type":43,"tag":111,"props":169,"children":170},{},[171,173,178],{"type":49,"value":172},"Core Pydantic AI usage -- building agents, adding tools, structured output, streaming, or testing (use ",{"type":43,"tag":58,"props":174,"children":176},{"className":175},[],[177],{"type":49,"value":91},{"type":49,"value":179},")",{"type":43,"tag":111,"props":181,"children":182},{},[183,185,190],{"type":49,"value":184},"Capabilities that ship in core ",{"type":43,"tag":58,"props":186,"children":188},{"className":187},[],[189],{"type":49,"value":63},{"type":49,"value":191},", such as web search, tool search, and thinking",{"type":43,"tag":111,"props":193,"children":194},{},[195,197,202,204,210],{"type":49,"value":196},"The Pydantic validation library on its own (",{"type":43,"tag":58,"props":198,"children":200},{"className":199},[],[201],{"type":49,"value":8},{"type":49,"value":203},"\u002F",{"type":43,"tag":58,"props":205,"children":207},{"className":206},[],[208],{"type":49,"value":209},"BaseModel",{"type":49,"value":211}," without agents)",{"type":43,"tag":95,"props":213,"children":215},{"id":214},"supported-capabilities",[216],{"type":49,"value":217},"Supported Capabilities",{"type":43,"tag":52,"props":219,"children":220},{},[221,226],{"type":43,"tag":58,"props":222,"children":224},{"className":223},[],[225],{"type":49,"value":128},{"type":49,"value":227}," has a full reference below; it is the flagship capability and the one this skill goes deep on.\nThe rest ship today and each has its own README with API and examples.",{"type":43,"tag":52,"props":229,"children":230},{},[231,233,239,241,247,249,254,255,261,262,268,270,276,278,283],{"type":49,"value":232},"Each capability lives in its own submodule and is imported from there\n(",{"type":43,"tag":58,"props":234,"children":236},{"className":235},[],[237],{"type":49,"value":238},"from pydantic_ai_harness.\u003Cmodule> import ...",{"type":49,"value":240},"). Capabilities are not importable from the top-level\n",{"type":43,"tag":58,"props":242,"children":244},{"className":243},[],[245],{"type":49,"value":246},"pydantic_ai_harness",{"type":49,"value":248}," package by design, so each one keeps its own optional dependencies isolated.\n",{"type":43,"tag":58,"props":250,"children":252},{"className":251},[],[253],{"type":49,"value":128},{"type":49,"value":122},{"type":43,"tag":58,"props":256,"children":258},{"className":257},[],[259],{"type":49,"value":260},"FileSystem",{"type":49,"value":122},{"type":43,"tag":58,"props":263,"children":265},{"className":264},[],[266],{"type":49,"value":267},"Shell",{"type":49,"value":269},", and ",{"type":43,"tag":58,"props":271,"children":273},{"className":272},[],[274],{"type":49,"value":275},"ManagedPrompt",{"type":49,"value":277}," also have top-level re-exports (importable directly\nfrom ",{"type":43,"tag":58,"props":279,"children":281},{"className":280},[],[282],{"type":49,"value":246},{"type":49,"value":284},").",{"type":43,"tag":52,"props":286,"children":287},{},[288],{"type":49,"value":289},"APIs are subject to change between releases; breaking changes ship deprecation warnings where practical.",{"type":43,"tag":291,"props":292,"children":293},"table",{},[294,318],{"type":43,"tag":295,"props":296,"children":297},"thead",{},[298],{"type":43,"tag":299,"props":300,"children":301},"tr",{},[302,308,313],{"type":43,"tag":303,"props":304,"children":305},"th",{},[306],{"type":49,"value":307},"Capability",{"type":43,"tag":303,"props":309,"children":310},{},[311],{"type":49,"value":312},"Module",{"type":43,"tag":303,"props":314,"children":315},{},[316],{"type":49,"value":317},"Description",{"type":43,"tag":319,"props":320,"children":321},"tbody",{},[322,365,391,417,443,469,495,521,558,584,610,636,670,696],{"type":43,"tag":299,"props":323,"children":324},{},[325,334,345],{"type":43,"tag":326,"props":327,"children":328},"td",{},[329],{"type":43,"tag":58,"props":330,"children":332},{"className":331},[],[333],{"type":49,"value":128},{"type":43,"tag":326,"props":335,"children":336},{},[337,343],{"type":43,"tag":58,"props":338,"children":340},{"className":339},[],[341],{"type":49,"value":342},"pydantic_ai_harness.code_mode",{"type":49,"value":344}," (also top-level)",{"type":43,"tag":326,"props":346,"children":347},{},[348,350,356,358],{"type":49,"value":349},"Wraps eligible tools into a single sandboxed ",{"type":43,"tag":58,"props":351,"children":353},{"className":352},[],[354],{"type":49,"value":355},"run_code",{"type":49,"value":357}," tool so the model orchestrates them in Python -- see ",{"type":43,"tag":359,"props":360,"children":362},"a",{"href":361},".\u002Freferences\u002FCODE-MODE.md",[363],{"type":49,"value":364},"Code Mode",{"type":43,"tag":299,"props":366,"children":367},{},[368,376,386],{"type":43,"tag":326,"props":369,"children":370},{},[371],{"type":43,"tag":58,"props":372,"children":374},{"className":373},[],[375],{"type":49,"value":260},{"type":43,"tag":326,"props":377,"children":378},{},[379,385],{"type":43,"tag":58,"props":380,"children":382},{"className":381},[],[383],{"type":49,"value":384},"pydantic_ai_harness.filesystem",{"type":49,"value":344},{"type":43,"tag":326,"props":387,"children":388},{},[389],{"type":49,"value":390},"Read, write, edit, and search files under a root directory, with traversal prevention",{"type":43,"tag":299,"props":392,"children":393},{},[394,402,412],{"type":43,"tag":326,"props":395,"children":396},{},[397],{"type":43,"tag":58,"props":398,"children":400},{"className":399},[],[401],{"type":49,"value":267},{"type":43,"tag":326,"props":403,"children":404},{},[405,411],{"type":43,"tag":58,"props":406,"children":408},{"className":407},[],[409],{"type":49,"value":410},"pydantic_ai_harness.shell",{"type":49,"value":344},{"type":43,"tag":326,"props":413,"children":414},{},[415],{"type":49,"value":416},"Run commands in a subprocess with allowlists, a default denylist, timeouts, and env masking",{"type":43,"tag":299,"props":418,"children":419},{},[420,428,438],{"type":43,"tag":326,"props":421,"children":422},{},[423],{"type":43,"tag":58,"props":424,"children":426},{"className":425},[],[427],{"type":49,"value":275},{"type":43,"tag":326,"props":429,"children":430},{},[431,437],{"type":43,"tag":58,"props":432,"children":434},{"className":433},[],[435],{"type":49,"value":436},"pydantic_ai_harness.logfire",{"type":49,"value":344},{"type":43,"tag":326,"props":439,"children":440},{},[441],{"type":49,"value":442},"Back an agent's instructions with a Logfire-managed prompt",{"type":43,"tag":299,"props":444,"children":445},{},[446,455,464],{"type":43,"tag":326,"props":447,"children":448},{},[449],{"type":43,"tag":58,"props":450,"children":452},{"className":451},[],[453],{"type":49,"value":454},"SubAgents",{"type":43,"tag":326,"props":456,"children":457},{},[458],{"type":43,"tag":58,"props":459,"children":461},{"className":460},[],[462],{"type":49,"value":463},"pydantic_ai_harness.subagents",{"type":43,"tag":326,"props":465,"children":466},{},[467],{"type":49,"value":468},"Delegate subtasks to specialized child agents",{"type":43,"tag":299,"props":470,"children":471},{},[472,481,490],{"type":43,"tag":326,"props":473,"children":474},{},[475],{"type":43,"tag":58,"props":476,"children":478},{"className":477},[],[479],{"type":49,"value":480},"DynamicWorkflow",{"type":43,"tag":326,"props":482,"children":483},{},[484],{"type":43,"tag":58,"props":485,"children":487},{"className":486},[],[488],{"type":49,"value":489},"pydantic_ai_harness.dynamic_workflow",{"type":43,"tag":326,"props":491,"children":492},{},[493],{"type":49,"value":494},"Orchestrate sub-agents from a model-written Python script",{"type":43,"tag":299,"props":496,"children":497},{},[498,507,516],{"type":43,"tag":326,"props":499,"children":500},{},[501],{"type":43,"tag":58,"props":502,"children":504},{"className":503},[],[505],{"type":49,"value":506},"Planning",{"type":43,"tag":326,"props":508,"children":509},{},[510],{"type":43,"tag":58,"props":511,"children":513},{"className":512},[],[514],{"type":49,"value":515},"pydantic_ai_harness.planning",{"type":43,"tag":326,"props":517,"children":518},{},[519],{"type":49,"value":520},"Break complex tasks into structured plans before execution",{"type":43,"tag":299,"props":522,"children":523},{},[524,544,553],{"type":43,"tag":326,"props":525,"children":526},{},[527,529,535,536,542],{"type":49,"value":528},"compaction family (",{"type":43,"tag":58,"props":530,"children":532},{"className":531},[],[533],{"type":49,"value":534},"SlidingWindowCompaction",{"type":49,"value":122},{"type":43,"tag":58,"props":537,"children":539},{"className":538},[],[540],{"type":49,"value":541},"SummarizingCompaction",{"type":49,"value":543},", ...)",{"type":43,"tag":326,"props":545,"children":546},{},[547],{"type":43,"tag":58,"props":548,"children":550},{"className":549},[],[551],{"type":49,"value":552},"pydantic_ai_harness.compaction",{"type":43,"tag":326,"props":554,"children":555},{},[556],{"type":49,"value":557},"Trim or summarize conversation history to stay within token limits",{"type":43,"tag":299,"props":559,"children":560},{},[561,570,579],{"type":43,"tag":326,"props":562,"children":563},{},[564],{"type":43,"tag":58,"props":565,"children":567},{"className":566},[],[568],{"type":49,"value":569},"ToolOutputLimits",{"type":43,"tag":326,"props":571,"children":572},{},[573],{"type":43,"tag":58,"props":574,"children":576},{"className":575},[],[577],{"type":49,"value":578},"pydantic_ai_harness.tool_output_limits",{"type":43,"tag":326,"props":580,"children":581},{},[582],{"type":49,"value":583},"Truncate, summarize, or spill large tool outputs",{"type":43,"tag":299,"props":585,"children":586},{},[587,596,605],{"type":43,"tag":326,"props":588,"children":589},{},[590],{"type":43,"tag":58,"props":591,"children":593},{"className":592},[],[594],{"type":49,"value":595},"RepoContext",{"type":43,"tag":326,"props":597,"children":598},{},[599],{"type":43,"tag":58,"props":600,"children":602},{"className":601},[],[603],{"type":49,"value":604},"pydantic_ai_harness.repo_context",{"type":43,"tag":326,"props":606,"children":607},{},[608],{"type":49,"value":609},"Auto-load CLAUDE.md\u002FAGENTS.md and repo structure",{"type":43,"tag":299,"props":611,"children":612},{},[613,622,631],{"type":43,"tag":326,"props":614,"children":615},{},[616],{"type":43,"tag":58,"props":617,"children":619},{"className":618},[],[620],{"type":49,"value":621},"StepPersistence",{"type":43,"tag":326,"props":623,"children":624},{},[625],{"type":43,"tag":58,"props":626,"children":628},{"className":627},[],[629],{"type":49,"value":630},"pydantic_ai_harness.step_persistence",{"type":43,"tag":326,"props":632,"children":633},{},[634],{"type":49,"value":635},"Save, restore, resume, and fork run state",{"type":43,"tag":299,"props":637,"children":638},{},[639,648,657],{"type":43,"tag":326,"props":640,"children":641},{},[642],{"type":43,"tag":58,"props":643,"children":645},{"className":644},[],[646],{"type":49,"value":647},"PydanticAIDocs",{"type":43,"tag":326,"props":649,"children":650},{},[651],{"type":43,"tag":58,"props":652,"children":654},{"className":653},[],[655],{"type":49,"value":656},"pydantic_ai_harness.pydantic_ai_docs",{"type":43,"tag":326,"props":658,"children":659},{},[660,662,668],{"type":49,"value":661},"On-demand ",{"type":43,"tag":58,"props":663,"children":665},{"className":664},[],[666],{"type":49,"value":667},"read_pyai_docs",{"type":49,"value":669}," tool for Pydantic AI docs",{"type":43,"tag":299,"props":671,"children":672},{},[673,682,691],{"type":43,"tag":326,"props":674,"children":675},{},[676],{"type":43,"tag":58,"props":677,"children":679},{"className":678},[],[680],{"type":49,"value":681},"CapabilityCreation",{"type":43,"tag":326,"props":683,"children":684},{},[685],{"type":43,"tag":58,"props":686,"children":688},{"className":687},[],[689],{"type":49,"value":690},"pydantic_ai_harness.capability_creation",{"type":43,"tag":326,"props":692,"children":693},{},[694],{"type":49,"value":695},"Let an agent author, validate, and load real capabilities at runtime",{"type":43,"tag":299,"props":697,"children":698},{},[699,704,713],{"type":43,"tag":326,"props":700,"children":701},{},[702],{"type":49,"value":703},"media externalization",{"type":43,"tag":326,"props":705,"children":706},{},[707],{"type":43,"tag":58,"props":708,"children":710},{"className":709},[],[711],{"type":49,"value":712},"pydantic_ai_harness.media",{"type":43,"tag":326,"props":714,"children":715},{},[716,718,724],{"type":49,"value":717},"Offload large ",{"type":43,"tag":58,"props":719,"children":721},{"className":720},[],[722],{"type":49,"value":723},"BinaryContent",{"type":49,"value":725}," to content-addressed stores",{"type":43,"tag":52,"props":727,"children":728},{},[729,731,737,739,745],{"type":49,"value":730},"Still experimental: an ACP server adapter, imported from ",{"type":43,"tag":58,"props":732,"children":734},{"className":733},[],[735],{"type":49,"value":736},"pydantic_ai_harness.experimental.acp",{"type":49,"value":738},". Importing it\nemits a ",{"type":43,"tag":58,"props":740,"children":742},{"className":741},[],[743],{"type":49,"value":744},"HarnessExperimentalWarning",{"type":49,"value":746},".",{"type":43,"tag":52,"props":748,"children":749},{},[750,752,760],{"type":49,"value":751},"The full, current list with links and status is in the\n",{"type":43,"tag":359,"props":753,"children":757},{"href":754,"rel":755},"https:\u002F\u002Fgithub.com\u002Fpydantic\u002Fpydantic-ai-harness#capability-matrix",[756],"nofollow",[758],{"type":49,"value":759},"capability matrix",{"type":49,"value":746},{"type":43,"tag":95,"props":762,"children":764},{"id":763},"install",[765],{"type":49,"value":766},"Install",{"type":43,"tag":768,"props":769,"children":774},"pre",{"className":770,"code":771,"language":772,"meta":773,"style":773},"language-bash shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","uv add pydantic-ai-harness\n","bash","",[775],{"type":43,"tag":58,"props":776,"children":777},{"__ignoreMap":773},[778],{"type":43,"tag":779,"props":780,"children":783},"span",{"class":781,"line":782},"line",1,[784,790,796],{"type":43,"tag":779,"props":785,"children":787},{"style":786},"--shiki-light:#E2931D;--shiki-default:#FFCB6B;--shiki-dark:#FFCB6B",[788],{"type":49,"value":789},"uv",{"type":43,"tag":779,"props":791,"children":793},{"style":792},"--shiki-light:#91B859;--shiki-default:#C3E88D;--shiki-dark:#C3E88D",[794],{"type":49,"value":795}," add",{"type":43,"tag":779,"props":797,"children":798},{"style":792},[799],{"type":49,"value":800}," pydantic-ai-harness\n",{"type":43,"tag":52,"props":802,"children":803},{},[804],{"type":49,"value":805},"Each capability declares its own extra. Code Mode needs the Monty sandbox:",{"type":43,"tag":768,"props":807,"children":809},{"className":770,"code":808,"language":772,"meta":773,"style":773},"uv add \"pydantic-ai-harness[codemode]\"   # `code-mode` is also accepted as an alias\n",[810],{"type":43,"tag":58,"props":811,"children":812},{"__ignoreMap":773},[813],{"type":43,"tag":779,"props":814,"children":815},{"class":781,"line":782},[816,820,824,830,835,840],{"type":43,"tag":779,"props":817,"children":818},{"style":786},[819],{"type":49,"value":789},{"type":43,"tag":779,"props":821,"children":822},{"style":792},[823],{"type":49,"value":795},{"type":43,"tag":779,"props":825,"children":827},{"style":826},"--shiki-light:#39ADB5;--shiki-default:#89DDFF;--shiki-dark:#89DDFF",[828],{"type":49,"value":829}," \"",{"type":43,"tag":779,"props":831,"children":832},{"style":792},[833],{"type":49,"value":834},"pydantic-ai-harness[codemode]",{"type":43,"tag":779,"props":836,"children":837},{"style":826},[838],{"type":49,"value":839},"\"",{"type":43,"tag":779,"props":841,"children":843},{"style":842},"--shiki-light:#90A4AE;--shiki-light-font-style:italic;--shiki-default:#546E7A;--shiki-default-font-style:italic;--shiki-dark:#676E95;--shiki-dark-font-style:italic",[844],{"type":49,"value":845},"   # `code-mode` is also accepted as an alias\n",{"type":43,"tag":52,"props":847,"children":848},{},[849,851,857],{"type":49,"value":850},"Requires Python 3.10+ and ",{"type":43,"tag":58,"props":852,"children":854},{"className":853},[],[855],{"type":49,"value":856},"pydantic-ai-slim>=2.18.0",{"type":49,"value":746},{"type":43,"tag":95,"props":859,"children":861},{"id":860},"quick-start",[862],{"type":49,"value":863},"Quick Start",{"type":43,"tag":52,"props":865,"children":866},{},[867,869,874,876,881],{"type":49,"value":868},"A harness capability is added to the agent like any other. Here ",{"type":43,"tag":58,"props":870,"children":872},{"className":871},[],[873],{"type":49,"value":128},{"type":49,"value":875}," wraps locally registered tools\ninto a single ",{"type":43,"tag":58,"props":877,"children":879},{"className":878},[],[880],{"type":49,"value":355},{"type":49,"value":882}," tool that the model drives with Python.",{"type":43,"tag":768,"props":884,"children":889},{"className":885,"code":886,"highlights":887,"language":16,"meta":773,"style":773},"language-python shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","from pydantic_ai import Agent\n\nfrom pydantic_ai_harness import CodeMode\n\nagent = Agent('anthropic:claude-sonnet-4-6', capabilities=[CodeMode()])\n\n\n@agent.tool_plain\ndef get_temperature_f(city: str) -> float:\n    return {'Paris': 68.0, 'Tokyo': 77.0}[city]\n\n\n@agent.tool_plain\ndef convert_temp(fahrenheit: float) -> float:\n    return round((fahrenheit - 32) * 5 \u002F 9, 1)\n\nresult = agent.run_sync(\n    'Compare the weather in Paris and Tokyo, and report both temperatures in Celsius.'\n)\nprint(result.output)\n#> Paris is 20.0 C and Tokyo is 25.0 C.\n",[888],0,[890],{"type":43,"tag":58,"props":891,"children":892},{"__ignoreMap":773},[893,901,911,919,927,936,944,952,961,970,979,987,995,1003,1012,1021,1029,1038,1047,1056,1065],{"type":43,"tag":779,"props":894,"children":895},{"class":781,"line":782},[896],{"type":43,"tag":779,"props":897,"children":898},{},[899],{"type":49,"value":900},"from pydantic_ai import Agent\n",{"type":43,"tag":779,"props":902,"children":904},{"class":781,"line":903},2,[905],{"type":43,"tag":779,"props":906,"children":908},{"emptyLinePlaceholder":907},true,[909],{"type":49,"value":910},"\n",{"type":43,"tag":779,"props":912,"children":913},{"class":781,"line":27},[914],{"type":43,"tag":779,"props":915,"children":916},{},[917],{"type":49,"value":918},"from pydantic_ai_harness import CodeMode\n",{"type":43,"tag":779,"props":920,"children":922},{"class":781,"line":921},4,[923],{"type":43,"tag":779,"props":924,"children":925},{"emptyLinePlaceholder":907},[926],{"type":49,"value":910},{"type":43,"tag":779,"props":928,"children":930},{"class":781,"line":929},5,[931],{"type":43,"tag":779,"props":932,"children":933},{},[934],{"type":49,"value":935},"agent = Agent('anthropic:claude-sonnet-4-6', capabilities=[CodeMode()])\n",{"type":43,"tag":779,"props":937,"children":939},{"class":781,"line":938},6,[940],{"type":43,"tag":779,"props":941,"children":942},{"emptyLinePlaceholder":907},[943],{"type":49,"value":910},{"type":43,"tag":779,"props":945,"children":947},{"class":781,"line":946},7,[948],{"type":43,"tag":779,"props":949,"children":950},{"emptyLinePlaceholder":907},[951],{"type":49,"value":910},{"type":43,"tag":779,"props":953,"children":955},{"class":781,"line":954},8,[956],{"type":43,"tag":779,"props":957,"children":958},{},[959],{"type":49,"value":960},"@agent.tool_plain\n",{"type":43,"tag":779,"props":962,"children":964},{"class":781,"line":963},9,[965],{"type":43,"tag":779,"props":966,"children":967},{},[968],{"type":49,"value":969},"def get_temperature_f(city: str) -> float:\n",{"type":43,"tag":779,"props":971,"children":973},{"class":781,"line":972},10,[974],{"type":43,"tag":779,"props":975,"children":976},{},[977],{"type":49,"value":978},"    return {'Paris': 68.0, 'Tokyo': 77.0}[city]\n",{"type":43,"tag":779,"props":980,"children":982},{"class":781,"line":981},11,[983],{"type":43,"tag":779,"props":984,"children":985},{"emptyLinePlaceholder":907},[986],{"type":49,"value":910},{"type":43,"tag":779,"props":988,"children":990},{"class":781,"line":989},12,[991],{"type":43,"tag":779,"props":992,"children":993},{"emptyLinePlaceholder":907},[994],{"type":49,"value":910},{"type":43,"tag":779,"props":996,"children":998},{"class":781,"line":997},13,[999],{"type":43,"tag":779,"props":1000,"children":1001},{},[1002],{"type":49,"value":960},{"type":43,"tag":779,"props":1004,"children":1006},{"class":781,"line":1005},14,[1007],{"type":43,"tag":779,"props":1008,"children":1009},{},[1010],{"type":49,"value":1011},"def convert_temp(fahrenheit: float) -> float:\n",{"type":43,"tag":779,"props":1013,"children":1015},{"class":781,"line":1014},15,[1016],{"type":43,"tag":779,"props":1017,"children":1018},{},[1019],{"type":49,"value":1020},"    return round((fahrenheit - 32) * 5 \u002F 9, 1)\n",{"type":43,"tag":779,"props":1022,"children":1024},{"class":781,"line":1023},16,[1025],{"type":43,"tag":779,"props":1026,"children":1027},{"emptyLinePlaceholder":907},[1028],{"type":49,"value":910},{"type":43,"tag":779,"props":1030,"children":1032},{"class":781,"line":1031},17,[1033],{"type":43,"tag":779,"props":1034,"children":1035},{},[1036],{"type":49,"value":1037},"result = agent.run_sync(\n",{"type":43,"tag":779,"props":1039,"children":1041},{"class":781,"line":1040},18,[1042],{"type":43,"tag":779,"props":1043,"children":1044},{},[1045],{"type":49,"value":1046},"    'Compare the weather in Paris and Tokyo, and report both temperatures in Celsius.'\n",{"type":43,"tag":779,"props":1048,"children":1050},{"class":781,"line":1049},19,[1051],{"type":43,"tag":779,"props":1052,"children":1053},{},[1054],{"type":49,"value":1055},")\n",{"type":43,"tag":779,"props":1057,"children":1059},{"class":781,"line":1058},20,[1060],{"type":43,"tag":779,"props":1061,"children":1062},{},[1063],{"type":49,"value":1064},"print(result.output)\n",{"type":43,"tag":779,"props":1066,"children":1068},{"class":781,"line":1067},21,[1069],{"type":43,"tag":779,"props":1070,"children":1071},{},[1072],{"type":49,"value":1073},"#> Paris is 20.0 C and Tokyo is 25.0 C.\n",{"type":43,"tag":52,"props":1075,"children":1076},{},[1077,1079,1084,1086,1091],{"type":49,"value":1078},"The model writes a single Python script that fetches both temperatures with ",{"type":43,"tag":58,"props":1080,"children":1082},{"className":1081},[],[1083],{"type":49,"value":146},{"type":49,"value":1085}," and then\nconverts them -- performing four tool calls across two dependent stages in one ",{"type":43,"tag":58,"props":1087,"children":1089},{"className":1088},[],[1090],{"type":49,"value":355},{"type":49,"value":1092}," invocation.",{"type":43,"tag":95,"props":1094,"children":1096},{"id":1095},"key-practices",[1097],{"type":49,"value":1098},"Key Practices",{"type":43,"tag":107,"props":1100,"children":1101},{},[1102,1119,1135],{"type":43,"tag":111,"props":1103,"children":1104},{},[1105,1110,1112,1117],{"type":43,"tag":158,"props":1106,"children":1107},{},[1108],{"type":49,"value":1109},"Confirm a harness capability is actually needed.",{"type":49,"value":1111}," If core Pydantic AI tools and capabilities are enough, use the ",{"type":43,"tag":58,"props":1113,"children":1115},{"className":1114},[],[1116],{"type":49,"value":91},{"type":49,"value":1118}," skill instead -- don't reach for the harness by default.",{"type":43,"tag":111,"props":1120,"children":1121},{},[1122,1127,1129,1133],{"type":43,"tag":158,"props":1123,"children":1124},{},[1125],{"type":49,"value":1126},"Read the reference before writing code.",{"type":49,"value":1128}," Each capability has its own configuration, constraints, and gotchas -- load the linked reference (e.g. ",{"type":43,"tag":359,"props":1130,"children":1131},{"href":361},[1132],{"type":49,"value":364},{"type":49,"value":1134},") first.",{"type":43,"tag":111,"props":1136,"children":1137},{},[1138,1143,1145,1150,1152,1157,1159,1165],{"type":43,"tag":158,"props":1139,"children":1140},{},[1141],{"type":49,"value":1142},"Install the capability's extra.",{"type":49,"value":1144}," Importing ",{"type":43,"tag":58,"props":1146,"children":1148},{"className":1147},[],[1149],{"type":49,"value":128},{"type":49,"value":1151}," without ",{"type":43,"tag":58,"props":1153,"children":1155},{"className":1154},[],[1156],{"type":49,"value":834},{"type":49,"value":1158}," raises an ",{"type":43,"tag":58,"props":1160,"children":1162},{"className":1161},[],[1163],{"type":49,"value":1164},"ImportError",{"type":49,"value":1166},"; the Monty sandbox is an optional dependency.",{"type":43,"tag":95,"props":1168,"children":1170},{"id":1169},"common-gotchas",[1171],{"type":49,"value":1172},"Common Gotchas",{"type":43,"tag":107,"props":1174,"children":1175},{},[1176,1227,1243],{"type":43,"tag":111,"props":1177,"children":1178},{},[1179,1196,1198,1203,1205,1211,1213,1218,1220,1226],{"type":43,"tag":158,"props":1180,"children":1181},{},[1182,1188,1190,1195],{"type":43,"tag":58,"props":1183,"children":1185},{"className":1184},[],[1186],{"type":49,"value":1187},"native=True",{"type":49,"value":1189}," tools bypass ",{"type":43,"tag":58,"props":1191,"children":1193},{"className":1192},[],[1194],{"type":49,"value":128},{"type":49,"value":746},{"type":49,"value":1197}," Provider-native MCP servers and web search execute server-side, so ",{"type":43,"tag":58,"props":1199,"children":1201},{"className":1200},[],[1202],{"type":49,"value":355},{"type":49,"value":1204}," never sees them. Use ",{"type":43,"tag":58,"props":1206,"children":1208},{"className":1207},[],[1209],{"type":49,"value":1210},"native=False",{"type":49,"value":1212}," for client-side dispatch that ",{"type":43,"tag":58,"props":1214,"children":1216},{"className":1215},[],[1217],{"type":49,"value":128},{"type":49,"value":1219}," can wrap, but do not treat a remote server as trusted or sandboxed; see the ",{"type":43,"tag":359,"props":1221,"children":1223},{"href":1222},".\u002Freferences\u002FCODE-MODE.md#sandbox-restrictions",[1224],{"type":49,"value":1225},"Code Mode trust boundary",{"type":49,"value":746},{"type":43,"tag":111,"props":1228,"children":1229},{},[1230,1235,1237,1241],{"type":43,"tag":158,"props":1231,"children":1232},{},[1233],{"type":49,"value":1234},"The Monty sandbox is a Python subset.",{"type":49,"value":1236}," It has no third-party imports and only a small stdlib allowlist -- read ",{"type":43,"tag":359,"props":1238,"children":1239},{"href":1222},[1240],{"type":49,"value":364},{"type":49,"value":1242}," before debugging generated code that fails to run.",{"type":43,"tag":111,"props":1244,"children":1245},{},[1246,1256,1258,1263],{"type":43,"tag":158,"props":1247,"children":1248},{},[1249,1254],{"type":43,"tag":58,"props":1250,"children":1252},{"className":1251},[],[1253],{"type":49,"value":128},{"type":49,"value":1255}," needs its extra.",{"type":49,"value":1257}," Install ",{"type":43,"tag":58,"props":1259,"children":1261},{"className":1260},[],[1262],{"type":49,"value":834},{"type":49,"value":1264},", not the bare package.",{"type":43,"tag":1266,"props":1267,"children":1268},"style",{},[1269],{"type":49,"value":1270},"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":1272,"total":938},[1273,1285,1304,1317,1327,1341],{"slug":91,"name":91,"fn":1274,"description":1275,"org":1276,"tags":1277,"stars":23,"repoUrl":24,"updatedAt":1284},"build AI agents with Pydantic AI","Build AI agents with Pydantic AI — tools, capabilities (including on-demand loading), structured output, streaming, testing, and multi-agent patterns. Use when the user mentions Pydantic AI, imports pydantic_ai, or asks to build an AI agent, add tools\u002Fcapabilities, defer capability loading, stream output, define agents from YAML, or test agent behavior.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1278,1279,1282,1283],{"name":18,"slug":19,"type":13},{"name":1280,"slug":1281,"type":13},"LLM","llm",{"name":9,"slug":8,"type":13},{"name":15,"slug":16,"type":13},"2026-07-27T06:05:58.6635",{"slug":1286,"name":1286,"fn":1287,"description":1288,"org":1289,"tags":1290,"stars":23,"repoUrl":24,"updatedAt":1303},"logfire-instrumentation","instrument applications with Pydantic Logfire","Add Pydantic Logfire observability to applications and send as much useful telemetry as possible. Use this skill whenever the user asks to \"add logfire\", \"instrument with logfire\", \"add observability\", \"add tracing\", \"configure logfire\", \"add monitoring\", \"set me up properly\", \"send as much data as would be useful\", \"maximize observability\", \"monitor my host\u002Finfrastructure\u002FKubernetes\", or mentions Logfire in any context. Supports Python, JavaScript\u002FTypeScript, and Rust, plus host\u002Finfrastructure metrics via the OpenTelemetry Collector. Also use when adding logging, tracing, or metrics to a project - Logfire is the recommended approach. Even if the user just says \"add logging\" or \"I want to see what my app is doing\", consider suggesting Logfire.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1291,1294,1297,1300],{"name":1292,"slug":1293,"type":13},"Engineering","engineering",{"name":1295,"slug":1296,"type":13},"Logfire","logfire",{"name":1298,"slug":1299,"type":13},"Monitoring","monitoring",{"name":1301,"slug":1302,"type":13},"Observability","observability","2026-07-16T06:00:51.155435",{"slug":1305,"name":1305,"fn":1306,"description":1307,"org":1308,"tags":1309,"stars":23,"repoUrl":24,"updatedAt":1316},"logfire-query","query and analyze Logfire telemetry","Query and analyze Logfire telemetry data — traces, logs, spans, metrics, summaries, and SQL results. Use this skill when the user asks to \"query logfire\", \"search traces\", \"find logs\", \"query data\", \"search spans\", \"look up errors in logfire\", \"get metrics from logfire\", \"analyze telemetry\", \"summarize errors\", \"find root cause\", or add Logfire querying capabilities to code. Do not use this skill for direct Logfire UI, browser, live-view, Explore-page, or link-opening requests; use logfire-ui instead. If \"show\" or \"view\" wording is ambiguous, ask whether the user wants a UI view or query analysis.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1310,1313,1314,1315],{"name":1311,"slug":1312,"type":13},"Data Analysis","data-analysis",{"name":1295,"slug":1296,"type":13},{"name":1298,"slug":1299,"type":13},{"name":1301,"slug":1302,"type":13},"2026-07-16T05:59:18.201951",{"slug":1318,"name":1318,"fn":1319,"description":1320,"org":1321,"tags":1322,"stars":23,"repoUrl":24,"updatedAt":1326},"logfire-ui","navigate Logfire project pages","Open or return Logfire project pages, live views, trace links, and Explore pages in the Codex browser without querying telemetry first. Use this skill when the user asks to \"open in Logfire\", \"show in the live view\", \"open Explore\", \"open the UI\", \"show in Codex\", \"use the browser\", \"give me a link\", or asks for a Logfire GUI\u002Fbrowser\u002Flive-view presentation of a project, time range, service, span, trace, log, or filter. If \"show\" or \"view\" wording is ambiguous, ask whether the user wants a UI view or query analysis.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1323,1324,1325],{"name":1295,"slug":1296,"type":13},{"name":1298,"slug":1299,"type":13},{"name":1301,"slug":1302,"type":13},"2026-07-16T05:59:17.66486",{"slug":8,"name":8,"fn":1328,"description":1329,"org":1330,"tags":1331,"stars":23,"repoUrl":24,"updatedAt":1340},"model and validate data with Pydantic","Pydantic is a Python data validation and serialization library, based on type hints. Use this skill whenever you need to do relatively complex data modeling using Pydantic, e.g. when adding constraints, defining a model hierarchy with subclasses, etc.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1332,1335,1336,1337],{"name":1333,"slug":1334,"type":13},"Data Modeling","data-modeling",{"name":9,"slug":8,"type":13},{"name":15,"slug":16,"type":13},{"name":1338,"slug":1339,"type":13},"Validation","validation","2026-07-23T06:05:52.745966",{"slug":4,"name":4,"fn":5,"description":6,"org":1342,"tags":1343,"stars":23,"repoUrl":24,"updatedAt":25},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1344,1345,1346,1347],{"name":18,"slug":19,"type":13},{"name":21,"slug":22,"type":13},{"name":9,"slug":8,"type":13},{"name":15,"slug":16,"type":13},{"items":1349,"total":938},[1350,1357,1364,1371,1377,1384],{"slug":91,"name":91,"fn":1274,"description":1275,"org":1351,"tags":1352,"stars":23,"repoUrl":24,"updatedAt":1284},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1353,1354,1355,1356],{"name":18,"slug":19,"type":13},{"name":1280,"slug":1281,"type":13},{"name":9,"slug":8,"type":13},{"name":15,"slug":16,"type":13},{"slug":1286,"name":1286,"fn":1287,"description":1288,"org":1358,"tags":1359,"stars":23,"repoUrl":24,"updatedAt":1303},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1360,1361,1362,1363],{"name":1292,"slug":1293,"type":13},{"name":1295,"slug":1296,"type":13},{"name":1298,"slug":1299,"type":13},{"name":1301,"slug":1302,"type":13},{"slug":1305,"name":1305,"fn":1306,"description":1307,"org":1365,"tags":1366,"stars":23,"repoUrl":24,"updatedAt":1316},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1367,1368,1369,1370],{"name":1311,"slug":1312,"type":13},{"name":1295,"slug":1296,"type":13},{"name":1298,"slug":1299,"type":13},{"name":1301,"slug":1302,"type":13},{"slug":1318,"name":1318,"fn":1319,"description":1320,"org":1372,"tags":1373,"stars":23,"repoUrl":24,"updatedAt":1326},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1374,1375,1376],{"name":1295,"slug":1296,"type":13},{"name":1298,"slug":1299,"type":13},{"name":1301,"slug":1302,"type":13},{"slug":8,"name":8,"fn":1328,"description":1329,"org":1378,"tags":1379,"stars":23,"repoUrl":24,"updatedAt":1340},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1380,1381,1382,1383],{"name":1333,"slug":1334,"type":13},{"name":9,"slug":8,"type":13},{"name":15,"slug":16,"type":13},{"name":1338,"slug":1339,"type":13},{"slug":4,"name":4,"fn":5,"description":6,"org":1385,"tags":1386,"stars":23,"repoUrl":24,"updatedAt":25},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1387,1388,1389,1390],{"name":18,"slug":19,"type":13},{"name":21,"slug":22,"type":13},{"name":9,"slug":8,"type":13},{"name":15,"slug":16,"type":13}]