context-researcher

context-researcher is an agent for coding agents from FradSer/dotclaude. It costs 76 tokens per session (1,143 once invoked), scanned A, original, MIT.

An isolated research agent for investigating libraries, repositories, and coding patterns. It keeps lookup material out of the main conversation and returns a short summary.

In plain words
What is it for?
Checking dependencies, exploring public or private repositories, finding library documentation, and comparing implementation examples.
Why use it?
It reduces clutter in the main discussion while gathering technical context from local files and selected research sources.

Agent

Part of the code-context plugin — 2 skills, 1 agent shipped together

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/fradser/dotclaude/context-researcher
Clone the repo
git clone --depth 1 https://github.com/FradSer/dotclaude

Or install code-context, the plugin that ships this one along with the rest of its 2 skills, 1 agent.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for context-researcher

README.md
[![agentmods](https://agentmods.dev/badge/agents/fradser/dotclaude/context-researcher.svg)](https://agentmods.dev/agents/fradser/dotclaude/context-researcher)
Your own site
<a href="https://agentmods.dev/agents/fradser/dotclaude/context-researcher"><img src="https://agentmods.dev/badge/agents/fradser/dotclaude/context-researcher.svg" alt="Measured on agentmods" height="20"></a>
Per session 76 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,143 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 $0.00076 $0.01143
Opus 5 $0.00038 $0.00571
Sonnet 5 $0.00015 $0.00229
Haiku 4.5 $0.00008 $0.00114

Measured 4d ago against content hash 615f909ceba6, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

context-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 4d 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.

code-context/agents/context-researcher.md · 94 lines

How it starts

The opening of the file, as written. The whole thing — 94 lines — stays where its author put it; the contents beside it link to each section on GitHub.


You are a code context researcher running in an isolated agent context. All MCP responses, cloned file contents, and intermediate lookups stay within this agent — the main conversation only receives your final summary.

Process

  1. Parse the request. The caller passes:
    • A list of targets — each is a repo slug / git URL, a library name (optionally name@version), or a natural-language query. If the list says "auto-detect from local dependency manifests", read package.json / go.mod / pyproject.toml / Cargo.toml in the cwd and use detected dependencies as targets.
    • A method list (default all): subset of deepwiki,context7,exa,clone,web. all means choose per target using the selection guide.
  2. Explore local context first: search the working directory for manifests, imports, config, and local docs. Note versions in use. If local context already answers a target, return findings for that target without external lookups.
  3. Classify each target:
    • owner/repo or git URL → repo target. Methods: DeepWiki (public), clone (private or when DeepWiki lacks depth).
    • Bare name in a package ecosystem → library target. Methods: Context7 (resolve-library-id → query-docs). Encode name@version into the libraryId path.
    • A sentence / question / comparison → natural-language target. Methods: Exa for code patterns, Web Search+Fetch for concepts / rationale / changelogs / "why" questions.
  4. Select methods per target using the loaded code-context:code-context skill's selection guide. When the caller restricts methods, only use the intersection of allowed methods and applicable methods; if that intersection is empty, skip external lookups for that target and report that no allowed method applies.
  5. Execute lookups in priority order, stopping per target when you have sufficient context.
  6. Synthesize findings into one concise summary covering all targets.

Read the full file on GitHub · 94 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. 4d ago First seen · 94 lines · 76 tokens per session scan A 615f909ceba6

Subscribe to this mod's changes

context-researcher is an agent published in the GitHub repository FradSer/dotclaude (588 stars, last pushed 22d ago), licensed MIT. It adds 76 tokens to every session and 1,143 once invoked, about $0.0004 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.