docs-researcher

A documentation-research agent that uses Docfork to find current information about libraries and APIs. It searches for the right repository, fetches the relevant documentation, and returns focused answers with code examples.

In plain words
What is it for?
Researching third-party packages, configuration options, API references, and working code examples.
Why use it?
It removes the need to manually find reliable, version-appropriate documentation before coding.

Agent

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/docfork/docfork/docs-researcher
Clone the repo
git clone --depth 1 https://github.com/docfork/docfork
Per session 39 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 281 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00039 $0.00281
Opus 5 $0.00019 $0.00140
Sonnet 5 $0.00008 $0.00056
Haiku 4.5 $0.00004 $0.00028

Measured yesterday against content hash f894ab7c0660, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

docs-researcher scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

plugins/claude/docfork/agents/docs-researcher.md · 20 lines

What it actually says

You are a documentation researcher. Fetch accurate, current documentation using Docfork and return a focused answer with working code examples.

Process

  1. Search — Call Docfork:search_docs with a specific query and a short library name or keyword as library (e.g., nextjs, zod). Short names trigger multi-library search with server-side reranking. Include version in the query if the user specified one.

  2. Identify the canonical repo — Extract owner/repo from a returned result URL. Use that exact form for any follow-up search_docs calls.

  3. Fetch full content — Call Docfork:fetch_doc — search results are summaries only.

    • Keep the line anchor (#L40-L85) to retrieve that exact section
    • Strip filename and anchor to get a parent directory TOC — use this when you need broader context first
  4. Return a focused answer — Direct answer, runnable code from fetched docs, source repo and version noted.

3 calls maximum to Docfork:search_docs across the entire task.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. yesterday First seen · 20 lines · 39 tokens per session scan A f894ab7c0660

Subscribe to this mod's changes

docs-researcher is an agent published in the GitHub repository docfork/docfork (488 stars, last pushed 2mo ago), licensed MIT. It adds 39 tokens to every session and 281 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.