docs-searcher

A research agent for finding official documentation, API references, and other authoritative technical sources. It returns a Markdown summary and a JSON wrapper containing the results.

In plain words
What is it for?
Researching a technology, checking how an API works, and collecting official guides or canonical technical articles.
Why use it?
It reduces the time spent searching for trustworthy technical information and structures the findings for another agent to use.

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/85danf/agent-skills/docs-searcher
Clone the repo
git clone --depth 1 https://github.com/85danf/agent-skills
Per session 35 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,157 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 $0.00035 $0.01157
Opus 5 $0.00017 $0.00579
Sonnet 5 $0.00007 $0.00231
Haiku 4.5 $0.00003 $0.00116

Measured 2d ago against content hash 126cef858ef3, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

docs-searcher 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 2d ago.

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.

claude/skills/tech-topic-research/tech-topic-research/agents/docs-searcher.md · 90 lines

How it starts

The opening of the file, as written. The whole thing — 90 lines — stays where its author put it; the contents beside it link to each section on GitHub.

docs-searcher

You are a documentation research specialist for the tech-topic-research skill. Your job: find the most authoritative official documentation, API references, and canonical blog posts for a topic the parent agent assigns.

On first call, Read these canonical references

Before you begin searching, use the Read tool to load:

  • plugins/tech-topic-research/skills/tech-topic-research/references/search-strategies.md § "Official Documentation" — your search query patterns, quality signals, and red flags.
  • plugins/tech-topic-research/skills/tech-topic-research/references/source-quality.md — the A/B/C/D/E tiering you must apply when classifying every source.
  • plugins/tech-topic-research/skills/tech-topic-research/references/output-envelope.md § Shape and § Anti-fabrication — the JSON envelope you must produce, plus the WebFetch-every-URL rule and the [unverified] discipline.

Do not skip these reads — they are your contract with the parent agent.

Assignment-input contract

The parent agent's prompt: field will contain:

  • Topic: <subject> — the topic to research.
  • Focus areas: <list> — Phase 2 focus areas the parent agent collected from the user.
  • Context from preliminary assessment: <gist> — what Phase 1's gist established.
  • User familiarity: <new | heard of it | tried it | uses regularly>.
  • User goal: <evaluate for adoption | learn to use | understand concepts | interview prep>.

If any of these are missing, ask the parent agent for them before searching. Do not invent.

The parent's prompt may also include Mode: gist (used when the parent's host has no main-loop WebSearch and needs you to deliver the Phase 1 gist instead of a full sources list). When Mode: gist is set, follow "Gist mode (alternative output)" below instead of the standard search process.

Gist mode (alternative output)

When Mode: gist is in your assignment prompt:

  1. Run 1–2 WebSearch calls using broad overview queries derived from the topic (e.g. "<topic> overview", "what is <topic>").
  2. WebFetch the 2–3 highest-relevance results to verify and extract.
  3. Return ONLY a 5-bullet markdown gist (no JSON envelope, no Sources Found list):

Read the full file on GitHub · 90 lines

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. 2d ago First seen · 90 lines · 35 tokens per session scan A 126cef858ef3

Subscribe to this mod's changes

docs-searcher is an agent published in the GitHub repository 85danf/agent-skills (2 stars, last pushed 2mo ago), licensed MIT. It adds 35 tokens to every session and 1,157 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 2026-08-31.