plugin
Open on GitHub
docvet — Claude Code plugin: Python docstring quality checks
Flags missing Raises, Yields, Attributes, Examples and cross-reference sections in Google-style Python docstrings while you type.
Install into Claude Code · Plugin marketplace
All methods and platforms ↓
/plugin marketplace add anthropics/claude-plugins-community
/plugin install docvet@claude-communityWhat docvet does
docvet is a CLI that audits Python docstrings, and this plugin wires it into Claude. It parses your code with AST and compares the structure against the prose: a function that raises ValueError without a Raises section gets flagged. The rule catalog covers 32 entries across five areas — docstring presence, section completeness, staleness from git history, griffe rendering warnings, and missing __init__.py files that keep a package invisible to mkdocs.
Findings show up inline. Required rules arrive as warnings, recommended ones as hints. Configuration sits under [tool.docvet] in pyproject.toml, where you also set fail-on and the freshness thresholds. Pre-commit hooks and a GitHub Action are included, so the same checks can run in CI.
Rendering checks need an extra package: pip install docvet[griffe]. Skip it and the griffe rules never run. If griffe appears in fail-on, docvet prints that to stderr and exits 1, which is a different outcome from a gate that actually found problems. Pass --no-fail-on-unavailable to warn and exit 0 instead.
What it helps with
- catch a function that raises an exception with no Raises section
- verify every signature parameter shows up under Args
- spot a docstring left behind after the signature or body changed
- check whether a package directory has
__init__.pyso mkdocs can find it - surface griffe parser warnings before the docs build
Example prompts for the agent
- Check the docstrings under
src/mypackage/and list the functions missing a Raises section - Run docvet check --all and tell me which findings are required and which are hints
- Why does docvet flag
parse_configinutils.py?
How to install docvet
Plugin marketplace
/plugin marketplace add anthropics/claude-plugins-community/plugin install docvet@claude-communityStep by step, with commands for each platform: Claude Code plugins: marketplaces, installing, and building your own
Questions about docvet
How do I install docvet in Claude Code?
From the plugin marketplace: run /plugin marketplace add anthropics/claude-plugins-community first, then /plugin install docvet@claude-community.
What does docvet do?
It checks Google-style Python docstrings for presence, section completeness, staleness signals from git history, griffe rendering warnings, and missing __init__.py files. Findings appear inline as you write code.
What is needed for the griffe rendering checks?
The optional griffe package, installed with pip install docvet[griffe]. Without it those rules do not run, and if griffe is listed in fail-on, docvet reports it on stderr and exits 1.
Source text
SKILL.md · MIT
[](https://github.com/Alberto-Codes/docvet/actions/workflows/ci.yml)
[](https://codecov.io/gh/Alberto-Codes/docvet)
[](https://pypi.org/project/docvet/)
[](https://pypi.org/project/docvet/)
[](https://github.com/Alberto-Codes/docvet/blob/main/LICENSE)
[](https://renovatebot.com)
[](https://github.com/astral-sh/ruff)
[](https://github.com/Alberto-Codes/docvet)
# docvet
**Better docstrings, better AI.**
## Why docvet?
docvet combines docstring presence metrics, structural checks, history-based staleness signals, and renderer diagnostics in one CLI, with editor and agent interfaces. These checks help reviewers find documentation gaps; they do not prove that prose describes runtime behavior correctly.
| Layer | Check | ruff | interrogate | pydoclint | **docvet** |
|-------|-------|------|-------------|-----------|------------|
| 1. Presence | "Does a docstring exist?" | Yes (D rules, opt-in) | Yes, with coverage metrics | Skips absent docstrings | **Yes, with coverage metrics** |
| 2. Style | "Is it formatted consistently?" | Yes (D rules, opt-in) | -- | Some parsing/style checks | Limited format checks |
| 3. Completeness | "Do sections and code structure agree?" | Partial (D rules; DOC rules in preview) | -- | Arguments, returns, yields, raises, attributes, type hints | **AST-based section and agreement checks** |
| 4. Accuracy | "What needs review for staleness?" | Structural checks above | -- | Structural checks above | **Git diff/blame heuristics; no semantic proof** |
| 5. Rendering | "Are there parser warnings?" | -- | -- | -- | **Griffe diagnostics** |
| 6. Visibility | "Is package discovery incomplete?" | -- | -- | -- | **Missing `__init__.py` checks** |
Comparison reviewed **2026-10-21** against primary documentation: Ruff's [D103 presence rule](https://docs.astral.sh/ruff/rules/undocumented-public-function/) and [DOC501 exception completeness rule](https://docs.astral.sh/ruff/rules/docstring-missing-exception/) (requires `--preview` and rule selection), [pydoclint's supported checks and styles](https://jsh9.github.io/pydoclint/), and [interrogate's coverage documentation](https://interrogate.readthedocs.io/en/latest/). Capabilities depend on enabled rules and configuration; `--` means no corresponding check is described in these sources.
**pydoclint** checks arguments, returns, yields, raises, class attributes, and type hints across Google, NumPy, and Sphinx styles. docvet's enrichment alone has 21 rules, including Raises, Yields, Receives, Warns, Attributes, Examples, cross-references, parameter agreement, and more. Add presence (coverage metrics + threshold enforcement), freshness (git diff/blame staleness signals), griffe rendering compatibility, and mkdocs coverage: 32 rules across 5 checks. Rule totals measure catalog entries, not equivalent coverage between tools.
Structural agreement checks compare documented elements with code structure. Freshness uses change history to suggest review: code can change without invalidating prose, and simultaneous code/docstring edits can still leave incorrect prose. Neither establishes semantic correctness.
**[Quickstart](#quickstart)** | **[GitHub Action](#github-action)** | **[Pre-commit](#pre-commit)** | **[Configuration](#configuration)** | **[AI Agent Integration](#ai-agent-integration)** | **[Docs](https://alberto-codes.github.io/docvet/)**
## What It Checks
**Presence** (existence) -- 2 rules:
`missing-docstring` `overload-has-docstring`
**Enrichment** (completeness) -- 21 rules:
`missing-raises` `missing-returns` `missing-yields` `missing-receives` `missing-warns` `missing-deprecation` `missing-param-in-docstring` `extra-param-in-docstring` `missing-other-parameters` `missing-attributes` `undocumented-init-params` `missing-typed-attributes` `missing-examples` `missing-cross-references` `extra-raises-in-docstring` `extra-yields-in-docstring` `extra-returns-in-docstring` `missing-return-type` `trivial-docstring` `prefer-fenced-code-blocks` `scaffold-incomplete`
**Freshness** (staleness signals) -- 5 rules:
`stale-signature` `stale-body` `stale-import` `stale-drift` `stale-age`
**Griffe** (rendering) -- 3 rules:
`griffe-unknown-param` `griffe-missing-type` `griffe-format-warning`
**Coverage** (visibility) -- 1 rule:
`missing-init`
## Quickstart
```bash
pip install docvet && docvet check --all
```
For optional griffe rendering checks:
```bash
pip install docvet[griffe]
```
Example output:
```
src/mypackage/helpers.py:1: missing-docstring Module has no docstring [required]
src/mypackage/utils.py:42: missing-raises Function 'parse_config' raises ValueError but has no Raises section [required]
src/mypackage/models.py:15: stale-signature Function 'process' signature changed but docstring not updated [required]
src/mypackage/api.py:1: missing-init Package directory missing __init__.py (invisible to mkdocs) [required]
```
## Configuration
Configure via `[tool.docvet]` in your `pyproject.toml`. All checks run and print findings. Checks listed in `fail-on` cause a non-zero exit code; unlisted checks are treated as warnings.
A check in `fail-on` that cannot run — most often `griffe` without the `docvet[griffe]` extra — never certified the gate you configured, so docvet reports it on stderr and exits 1. With `--format json` the `run` object reports `status: "unavailable"` and an `unavailable_checks` array, which is what distinguishes it from a gate that found problems. Set `fail-on-unavailable = false` (or pass `--no-fail-on-unavailable`) to warn and exit 0 instead. Unavailable checks that nothing gates on stay a quiet skip — though `fail-on` membership is not the only thing that gates: a `min-coverage` floor gates `presence` without naming it there.
```toml
[tool.docvet]
exclude = ["tests", "scripts"]
fail-on = ["griffe", "coverage"]
[tool.docvet.freshness]
drift-threshold = 30
age-threshold = 90
```
## Pre-commit
Add to your `.pre-commit-config.yaml`:
```yaml
repos:
- repo: https://github.com/Alberto-Codes/docvet
rev: v3.0.0
hooks:
- id: docvet
```
For griffe rendering checks, add the optional dependency:
```yaml
repos:
- repo: https://github.com/Alberto-Codes/docvet
rev: v3.0.0
hooks:
- id: docvet
additional_dependencies: [griffe]
```
## GitHub Action
Add docvet to your GitHub Actions workflow — findings appear as inline annotations on your PR:
```yaml
- uses: Alberto-Codes/[email protected]
```
Select specific checks or pin a version:
```yaml
- uses: Alberto-Codes/[email protected]
with:
checks: 'enrichment,freshness'
docvet-version: '3.0.0'
python-version: '3.13'
```
The griffe rendering check needs no setup: the action installs `docvet[griffe]`, pinned or not, so every check the `checks` input offers is available. Note that docvet releases before 1.7.0 declare that extra without an upper bound, so pinning one installs whatever griffe publishes at the time rather than a version docvet was released against.
> **Behavior change — this can turn a passing build red.**
>
> Earlier releases installed plain `docvet`, so the griffe check was skipped and contributed zero findings. It now runs. `determine_run_outcome` (`src/docvet/reporting.py`) returns exit code 1 as soon as any check listed in `fail-on` reports findings, so if your `pyproject.toml` has `griffe` in `[tool.docvet] fail-on`, your build goes from green to failing with no change on your side. This repository's own `ci.yml` docvet job is exactly such a consumer.
>
> These are not new problems — it is the check finally running on docstrings that were always broken. To get back to green, fix the griffe findings or remove `griffe` from `fail-on`.
## AI Agent Integration
For tool-specific integration snippets, see the [full AI Agent Integration guide](https://alberto-codes.github.io/docvet/ai-integration/).
Add docvet to your AI coding workflow. Drop this into your `CLAUDE.md`, `.cursorrules`, or agent configuration:
```markdown
## Docstring Quality
After modifying Python functions, classes, or modules, run `docvet check` and fix all findings before committing.
```
Recommended `pyproject.toml` configuration:
```toml
[tool.docvet]
fail-on = ["enrichment", "freshness", "coverage", "griffe"]
```
### Subcommand Quick Reference
| Command | Description |
|---------|-------------|
| `docvet check` | Run all enabled checks (default: git diff files) |
| `docvet check --all` | Run all checks on entire codebase |
| `docvet check --staged` | Run all checks on staged files only |
| `docvet presence` | Check for missing docstrings with coverage metrics |
| `docvet enrichment` | Check for missing docstring sections |
| `docvet freshness` | Flag potentially stale docstrings via git |
| `docvet freshness --mode drift` | Review potential staleness via git blame |
| `docvet coverage` | Find files invisible to mkdocs |
| `docvet griffe` | Check mkdocs rendering compatibility |
| `docvet fix` | Scaffold missing docstring sections |
| `docvet fix --dry-run` | Preview scaffolding changes without writing files |
| `docvet config` | Show effective configuration with source annotations |
| `docvet lsp` | Start LSP server for real-time editor diagnostics |
| `docvet mcp` | Start MCP server for AI agent integration |
## Better Docstrings, Better AI
AI coding agents rely on docstrings as context when generating and modifying code. Agents modify code but often leave docstrings stale, and research shows stale or incorrect documentation is actively harmful, worse than no docs at all:
- Incorrect docs [degrade LLM task success by 22.6 percentage points](https://arxiv.org/abs/2404.03114)
- Comment density [improves code generation by 40-54%](https://arxiv.org/abs/2402.13013)
- Misleading comments [reduce LLM fault localization accuracy to 24.55%](https://arxiv.org/abs/2504.04372)
- Performance [drops substantially without docstrings](https://arxiv.org/abs/2508.09537)
As the [2025 DORA report](https://cloud.google.com/resources/content/2025-dora-ai-assisted-software-development-report) puts it: "AI doesn't fix a team; it amplifies what's already there." The [only signal correlating with AI productivity is code quality](https://stackoverflow.blog/2026/02/04/code-smells-for-ai-agents-q-and-a-with-eno-reyes-of-factory).
docvet's freshness heuristics flag potential staleness for review, and its enrichment rules check supported docstring sections against code structure. Run `docvet check` in your CI, pre-commit hooks, or agent toolchain.
## Badge
Add a badge to your project to show your docs are vetted:
```markdown
[](https://github.com/Alberto-Codes/docvet)
```
## Used By
Are you using docvet? Open a pull request to add your project here.
## License
MIT -- see [LICENSE](https://github.com/Alberto-Codes/docvet/blob/main/LICENSE) for details.
mcp-name: io.github.Alberto-Codes/docvet
© Alberto-Codes · MIT · Alberto-Codes/docvet
Similar plugins
All plugins for Writing and docs →
claude-api
official This plugin adds Claude API and SDK guidance for building LLM-powered applications. Install it when your project needs a Claude integration.
★ 184,013repo stars
document-skills
official Document processing suite for Excel, Word, PowerPoint, and PDF files. Equips the assistant to inspect and generate common document formats.
★ 184,013repo stars
project-artifact
official This plugin builds a living project status page and publishes it as a claude.ai artifact. Later refreshes reread project sources, keep the same URL, and report changes.
★ 44,770repo stars
context7
official Connects Claude Code to Context7's hosted MCP server to fetch up-to-date library docs and code examples straight from source repositories.
★ 44,770repo stars
claude-md-management
official Audits CLAUDE.md files against your codebase and records lessons learned during work sessions to keep project memory up to date.
★ 44,770repo stars
visual-explainer Turn architecture explanations, diffs, and requirement comparisons into styled HTML pages that open in a browser. Useful when terminal diagrams become hard to read.
★ 10,644repo stars
