Owner

pydantic

17 indexed repositories · View on GitHub

pydantic
Data validation using Python type hints

Python · 28826 stars
pydantic-ai
How Python does AI. Agents, realtime voice, image generation, embeddings. Every model, every interface, typed end to end.

Python · 20039 stars
monty
A minimal, secure Python interpreter written in Rust for use by AI

Rust · 8238 stars
logfire
AI observability platform for production LLM and agent systems.

Python · 4480 stars
pydantic-settings
Settings management using pydantic

Python · 1457 stars
httpx2
A next generation HTTP client for Python. 🦋

Python · 1453 stars
pydantic-ai-harness
Your agent's favorite harness, built on Pydantic AI

Python · 903 stars
jiter
Fast iterable JSON parser.

Rust · 550 stars
genai-prices
Calculate prices for calling LLM inference APIs.

Python · 376 stars
pydantic-extra-types
Extra Pydantic types.

Python · 335 stars
speedate
Fast and simple datetime, date, time and duration parsing for rust.

Rust · 252 stars
pytest-examples
Pytest plugin for testing examples in docstrings and markdown files.

Python · 178 stars
skills

Python · 135 stars
logfire-rust

Rust · 77 stars
logfire-js
JavaScript and TypeScript SDK for Logfire

TypeScript · 72 stars
typing-inspection
Runtime typing introspection tools

Python · 67 stars
ai-chat-ui
Web UI for Pydantic AI agents, built over the Vercel AI protocol

TypeScript · 65 stars

Beginner friendly Everything

Open All

DiskRetryer treats a non-retryable 4xx as a successful delivery and deletes the queued export Open

Difficulty 5/5 Over a week Newbie friendliness 35/100

pydantic/logfire#2443 · Opened Sep 19, 2026
An uninstrumented delegate's usage is credited to the caller's agent run span Open

pydanty:is-working

Difficulty 4/5 3-5 days Newbie friendliness 52/100

pydantic/pydantic-ai#8551 · Opened Sep 19, 2026
A provider profile for an endpoint with no key? Open

p:4-low pydanty:feature

Difficulty 4/5 3-5 days Newbie friendliness 48/100

pydantic/pydantic-ai#8550 · 1 comment · Opened Sep 19, 2026
Close `MCPToolset` stdio processes under outer AnyIO cancellation Open

pydanty:is-working

Difficulty 4/5 3-5 days Newbie friendliness 64/100

pydantic/pydantic-ai#8548 · 2 comments · Opened Sep 19, 2026
Complete the remaining `CLAI2` local usability checklist Open

agent-feature enhancement pydanty:is-working

pydantic/pydantic-ai-harness#966 · 2 comments · 1 reaction · 1 assignee · Opened Sep 19, 2026
Root logger level gets stuck at ERROR forever if griffe raises while parsing a tool docstring Open

p:2-high pydanty:bug

Difficulty 2/5 1-3 hours Newbie friendliness 88/100

pydantic/pydantic-ai#8547 · 2 comments · Opened Sep 19, 2026
MontyRepl exhausts the u16 name id space after ~65k inputs: each <python-input-N> filename is interned permanently Open

Difficulty 4/5 3-5 days Newbie friendliness 48/100

pydantic/monty#895 · Opened Sep 19, 2026
No cassette covers a Jev run where a filled tool call and a union meet, and `state` changes shape mid-run Open

Difficulty 3/5 1-2 days Newbie friendliness 72/100

pydantic/pydantic-ai#8539 · Opened Sep 19, 2026
`None` as a Jev union member asks for a docstring it cannot have, then demands a fallback model for a route with nothing to fill Open

Difficulty 3/5 1-2 days Newbie friendliness 70/100

pydantic/pydantic-ai#8538 · 2 comments · Opened Sep 19, 2026
The latest-versions canary is failing: ag-ui-protocol dropped ThinkingEndEvent, and gpt-audio-mini is missing from known model names Open

Difficulty 3/5 1-2 days Newbie friendliness 68/100

pydantic/pydantic-ai#8535 · 1 comment · Opened Sep 19, 2026
Split `UserError` so a refusal a better model could answer can reach a `FallbackModel` Open

Difficulty 5/5 Over a week Newbie friendliness 35/100

pydantic/pydantic-ai#8532 · Opened Sep 19, 2026
Bump the gh-aw engine pin to 0.32.0 Open

chore p:4-low pydanty:meta

pydantic/pydantic-ai-harness#956 · 1 comment · 1 assignee · Opened Sep 19, 2026
An output function sees its own unanswered output tool call in `ctx.messages`, but only under tool output Open

Difficulty 5/5 Over a week Newbie friendliness 45/100

pydantic/pydantic-ai#8531 · 1 comment · Opened Sep 19, 2026
Flaky: `test_supervisor_killed_mid_command_returns_stale_status` races supervisor reaping Open

pydantic/pydantic-ai-harness#954 · 1 assignee · Opened Sep 19, 2026
A `functools.partial` passed to `ToolOutput` is treated as a type, adding a parameter that should not exist Open

Difficulty 3/5 1-2 days Newbie friendliness 78/100

pydantic/pydantic-ai#8527 · 1 comment · Opened Sep 19, 2026
Compact a history by scoring what is still relevant, instead of summarising it Open

Difficulty 5/5 Over a week Newbie friendliness 25/100

pydantic/pydantic-ai#8526 · Opened Sep 19, 2026
OpenAIChatModel auto send-back should backfill the detected thinking field Open

bug p:2-high pydanty:bug

pydantic/pydantic-ai#8525 · 6 comments · 1 assignee · Opened Sep 19, 2026
Describe a field by its docstring consistently, wherever the type is used Open

Difficulty 5/5 Over a week Newbie friendliness 38/100

pydantic/pydantic-ai#8522 · 1 comment · Opened Sep 19, 2026
Let an evaluator answer several measures in one model request Open

Difficulty 5/5 Over a week Newbie friendliness 35/100

pydantic/pydantic-ai#8519 · Opened Sep 18, 2026
Measure output modes per model before changing any `ModelProfile` default Open

Difficulty 5/5 Over a week Newbie friendliness 35/100

pydantic/pydantic-ai#8515 · Opened Sep 18, 2026
Record the capabilities and toolsets in use on agent run spans Open

feature

Difficulty 4/5 3-5 days Newbie friendliness 48/100

pydantic/pydantic-ai#8513 · Opened Sep 18, 2026
[aw] CI Review hit engine rate limit (HTTP 429) Open

agentic-workflows

Difficulty 4/5 3-5 days Newbie friendliness 25/100

pydantic/pydantic-ai#8511 · 2 comments · Opened Sep 18, 2026
[aw] CI Review produced no safe outputs Open

agentic-workflows pydanty:is-working

Difficulty 4/5 3-5 days Newbie friendliness 45/100

pydantic/pydantic-ai#8510 · 2 comments · Opened Sep 18, 2026
Host security capabilities do not reach sub-agent runs, so `delegate_task` bypasses parent-level tool policy Open

pydantic/pydantic-ai-harness#947 · 1 comment · 1 assignee · Opened Sep 18, 2026
[aw] CI Review failed Open

agentic-workflows pydanty:is-working

Difficulty 3/5 1-2 days Newbie friendliness 35/100

pydantic/pydantic-ai#8506 · 8 comments · Opened Sep 18, 2026
`Coder`'s sub-agent delegate inherits `repo_context=False` and loses the repository instructions Open

pydantic/pydantic-ai-harness#945 · 1 assignee · Opened Sep 18, 2026
Restore sub-agent delegation to the default `Coder`, behind a flag Open

pydantic/pydantic-ai-harness#944 · 1 assignee · Opened Sep 18, 2026
pydantic.v1._hypothesis_plugin cannot be imported under pydantic 2 Open

Difficulty 2/5 1-3 hours Newbie friendliness 85/100

pydantic/pydantic#13835 · 3 comments · Opened Sep 18, 2026
Track shipped CLAI 2 features and implementation commits Open

chore p:4-low pydanty:meta

pydantic/pydantic-ai-harness#942 · 1 comment · 1 assignee · Opened Sep 18, 2026
`create_model`'s `__validators__` annotation excludes the descriptor proxy returned by `model_validator` Open

bug V2 pending

Difficulty 3/5 1-2 days Newbie friendliness 68/100

pydantic/pydantic#13834 · 3 comments · Opened Sep 18, 2026
Field(init=False) is not reflected in dataclass metadata Open

Difficulty 3/5 1-2 days Newbie friendliness 65/100

pydantic/pydantic#13833 · Opened Sep 18, 2026
A model that makes several API calls in one `request()` cannot report them in `usage.requests` Open

feature

Difficulty 5/5 Over a week Newbie friendliness 35/100

pydantic/pydantic-ai#8498 · 2 comments · Opened Sep 18, 2026
HTTP/2 connection deadlocks when the peer lowers MAX_CONCURRENT_STREAMS below the in-flight stream count Open

Difficulty 4/5 3-5 days Newbie friendliness 55/100

pydantic/httpx2#1216 · 1 comment · Opened Sep 18, 2026
Add a typed `ModelRouter` capability Open

pydantic/pydantic-ai-harness#938 · 1 assignee · Opened Sep 18, 2026
`Path.read_text()` accepts no arguments, while `open(encoding=...)` works Open

Difficulty 3/5 1-2 days Newbie friendliness 73/100

pydantic/monty#893 · Opened Sep 18, 2026
Let `Advisor` take an `output_type`, so a model that does not write text can be consulted Open

pydantic/pydantic-ai-harness#922 · 1 assignee · Opened Sep 18, 2026
Server-initiated sampling and elicitation are unreachable on FastMCP 4 Open

bug

Difficulty 4/5 3-5 days Newbie friendliness 52/100

pydantic/pydantic-ai#8477 · 1 comment · Opened Sep 18, 2026
Usage a capability adds to `ctx.usage` inside a durable operation is credited on replay but not on a first run Open

bug

Difficulty 5/5 Over a week Newbie friendliness 35/100

pydantic/pydantic-ai#8476 · 1 comment · Opened Sep 18, 2026
Separate effective tool definitions from tool execution state Open

refactor tools

pydantic/pydantic-ai#8474 · 1 assignee · Opened Sep 18, 2026
Run an agent on a model that only answers questions: what `TypeSafeModel` (Jev) still needs from the framework Open

Difficulty 5/5 Over a week Newbie friendliness 25/100

pydantic/pydantic-ai#8472 · 18 comments · 1 reaction · Opened Sep 18, 2026
Rename `gen_ai.aggregated_usage.*`, which no longer aggregates Open

feature

Difficulty 5/5 Over a week Newbie friendliness 35/100

pydantic/pydantic-ai#8465 · Opened Sep 18, 2026
A `Model` cannot tell messages from `message_history` apart from the ones the agent just built Open

feature

Difficulty 5/5 Over a week Newbie friendliness 35/100

pydantic/pydantic-ai#8464 · Opened Sep 17, 2026
A text-like file that is not valid UTF-8 raises `UnicodeDecodeError` instead of being read or refused Open

bug

Difficulty 4/5 3-5 days Newbie friendliness 48/100

pydantic/pydantic-ai#8452 · 1 comment · Opened Sep 17, 2026
Dataset SDK rejects organization-wide API keys and provides no project selector Open

Difficulty 3/5 1-2 days Newbie friendliness 55/100

pydantic/logfire#2434 · 1 comment · Opened Sep 17, 2026
Integrate `clai2` natively with Agent of Empires through a CLAI plugin Open

pydantic/pydantic-ai-harness#915 · 1 assignee · Opened Sep 17, 2026
External futures can't be awaited across feeds within the same session Open

Difficulty 3/5 1-2 days Newbie friendliness 74/100

pydantic/monty#879 · 1 reaction · Opened Sep 17, 2026
[aw] Triage pilot (pydantic-ai engine) failed Open

agentic-workflows

Difficulty 3/5 1-2 days Newbie friendliness 45/100

pydantic/genai-prices#702 · 1 comment · Opened Sep 17, 2026
Groq's gpt-oss-safeguard-20b bills at the gpt-oss-120b rate, about double its real price Open

pydanty:is-working

Difficulty 2/5 1-3 hours Newbie friendliness 90/100

pydantic/genai-prices#701 · Opened Sep 17, 2026
Run the latest-versions canary on a nightly schedule, not only on tag refs Open

Difficulty 3/5 1-2 days Newbie friendliness 68/100

pydantic/pydantic-ai#8440 · 1 comment · Opened Sep 17, 2026
[aw] CI Review failed Open

agentic-workflows

Difficulty 4/5 3-5 days Newbie friendliness 25/100

pydantic/pydantic-ai#8435 · 20 comments · Opened Sep 17, 2026

Showing the newest 100

This page lists what was indexed most recently. Advanced filter has the whole inventory, narrowed by language, difficulty and how long a task takes.

Open advanced filter

pydantic

Showing the newest 100

Get new issues in your inbox