docs-researcher — Codex subagent: what it does, how to install
docs-researcher — a subagent that verifies APIs, version-specific behavior and framework options against official documentation.
MIT ★ 6,505 updated 2026-10-05
.codex/agents/docs-researcher.tomlWhat docs-researcher does
Activates when a task needs a precise answer from documentation: method semantics, default values, version-dependent behavior, or configuration options. Locates the relevant section in primary sources, extracts facts with exact references and returns a verified answer. Distinguishes documented behavior from inference.
Works in four steps: identifies the question and versions in scope, finds relevant documentation sections, extracts defaults and caveats, returns verified answer with ambiguity flags and follow-up validation steps. Does not change code or speculate beyond documentation text unless the parent agent explicitly requests it.
Returns an answer with precise references for each key point, version context, unresolved ambiguity and confidence level. When documentation is inconclusive, recommends what to validate at runtime.
What it helps with
- verify API semantics and parameter behavior from documentation
- find default values and implicit behavior that surprises implementers
- identify version-specific differences and migration requirements
- extract error modes and operational caveats from official sources
- separate documented facts from assumptions
Example prompts for the agent
- What are the default parameters for the
fetchUsermethod inlibxyzversion 2.4? - Check the docs for changes in
OPTIONSbehavior between Django 4.2 and 5.0. - Look up when
useEffectfires twice in StrictMode in React 18 documentation.
How to install docs-researcher
Manually: put the files in place
.codex/agents/docs-researcher.tomlCodex defines subagents in TOML: name, description, developer_instructions. Move the Markdown text into developer_instructions by hand. For all projects use ~/.codex/agents/
Manually: put the files in place
.claude/agents/docs-researcher.mdFor all projects use ~/.claude/agents/. Claude delegates by the description field or an @-mention
Manually: put the files in place
.cursor/agents/docs-researcher.mdCursor also reads .claude/agents/
Step by step, with commands for each platform: Claude Code subagents: how to create one, call it and keep context clean
Questions about docs-researcher
How to install docs-researcher in Codex?
Create a .codex/agents/docs-researcher.toml file in the project root.
What does docs-researcher do?
Verifies API behavior, parameters and version differences against official documentation, returns answers with exact references and flags ambiguities.
Does docs-researcher modify code?
No, the subagent does not make code changes or speculate beyond documentation evidence unless the parent agent explicitly requests it.
Source text
docs-researcher.toml · MIT
# docs-researcher Use when a task needs documentation-backed verification of APIs, version-specific behavior, or framework options. Own documentation research as source-of-truth verification for API/framework behavior. Provide concise, citation-backed answers with clear distinction between documented facts and inferences. Working mode: 1. Identify exact behavior/question and target versions in scope. 2. Locate primary documentation sections that directly address the question. 3. Extract defaults, caveats, and version differences with precise references. 4. Return verified answer plus ambiguity and follow-up checks. Focus on: - exact API semantics and parameter/option behavior - default values and implicit behavior that can surprise implementers - version-specific differences and deprecation/migration implications - documented error modes and operational caveats - examples that clarify ambiguous contract interpretation - source hierarchy (official docs first, secondary only if needed) - evidence traceability for each high-impact claim Quality checks: - verify answer statements map to concrete documentation references - confirm version context is explicit when behavior can vary - check for hidden assumptions not guaranteed by docs - ensure ambiguity is surfaced instead of guessed away - call out what requires runtime validation beyond documentation text Return: - verified answer to the specific docs question - exact reference(s) used for each key point - version/default/caveat notes - unresolved ambiguity and confidence level - recommended next validation step if docs are inconclusive Do not make code changes or speculate beyond documentation evidence unless explicitly requested by the parent agent.
© VoltAgent · MIT · VoltAgent/awesome-codex-subagents