agents-council: Skill for Claude Code

.agents/skills/context-hunter/SKILL.md

context-hunter is a skill for Claude Code, Codex from MrLesk/agents-council. It costs 39 tokens per session (1,157 once invoked), scanned A, original, MIT.

An MCP server for pymupdf4llm, a Python package for working with PDF documents for large-language-model tasks. The input does not describe the server's specific commands or supported operations.

In plain words
What is it for?
It is intended to provide access to pymupdf4llm through an MCP server, but the concrete jobs it supports are not specified.
Why use it?
The available information is too limited to say what problem this add-on removes.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: installed under .agents/ (shared by several agents).

This is MrLesk/agents-council's own configuration. It tells Claude Code and Codex how to work on agents-council itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything agents-council configures →

Reuse

Borrowing it

Nothing to install: this file belongs to MrLesk/agents-council. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/MrLesk/agents-council/main/.agents/skills/context-hunter/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/MrLesk/agents-council

Made for: Claude Code, Codex.

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-hunter

README.md
[![agentmods](https://agentmods.dev/badge/skills/mrlesk/agents-council/context-hunter.svg)](https://agentmods.dev/skills/mrlesk/agents-council/context-hunter)
Your own site
<a href="https://agentmods.dev/skills/mrlesk/agents-council/context-hunter"><img src="https://agentmods.dev/badge/skills/mrlesk/agents-council/context-hunter.svg" alt="Measured on agentmods" height="20"></a>
Per session 39 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,157 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00039 $0.01157
Opus 5 $0.00019 $0.00579
Sonnet 5 $0.00008 $0.00231
Haiku 4.5 $0.00004 $0.00116

Measured 8d ago against content hash 8f634b3adef2, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

context-hunter 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 8d 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.

.agents/skills/context-hunter/SKILL.md · 142 lines

How it starts

The opening of the file, as written. The whole thing — 142 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Context Hunter

Before writing code, run a focused discovery loop. Do not load everything. Find the right files.

Complexity Gate

Classify task complexity first:

  • L0 (trivial): typos, renames, copy-only edits, obvious single-line fixes with no behavior change.
  • L1 (moderate): behavior changes in one bounded area.
  • L2 (high-risk): cross-module changes, data semantics, refactors, architecture-impacting work.

Output by level:

  • L0: no context brief, proceed directly.
  • L1: write a micro-brief.
  • L2: write a full context brief.

Re-evaluate level during discovery and implementation. If new evidence shows higher complexity than initially classified, upgrade the level and apply the stricter workflow.

Core Behavior

Act like a senior engineer who asks the next useful question:

  1. Assess completeness: Check whether the request omits expected concerns seen in analogous code.
  2. Discover selectively: Read the minimum set of relevant files.
  3. Validate assumptions: Confirm with tests/config/history.
  4. Synthesize: Capture findings before coding for L1/L2.

Discovery Workflow (Before Coding)

1) Assess Request Completeness

Ask: "What is likely missing?"

Examples:

  • Similar endpoints include auth/validation. Is that expected here?
  • This area uses soft-delete semantics. Should this operation follow that?
  • Similar flows emit telemetry/error states. Should this change do the same?
  • Existing module boundaries suggest a different placement. Is current request still correct?

2) Run Targeted Discovery

Prioritize these in order:

  1. Find analogous implementations and copy their structure.
  2. Trace data flow for similar features end-to-end.
  3. Identify reusable utilities before creating new helpers.
  4. Inspect nearby tests to infer team priorities and edge cases.
  5. Read recent commits in the same area for current direction.

Portable discovery actions:

  • Search for feature/domain terms in relevant directories.
  • Enumerate nearby files in the affected area.
  • Inspect recent change history for touched paths.
  • Run targeted validation first, then broader project checks as needed.

Read the full file on GitHub · 142 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. 8d ago First seen · 142 lines · 39 tokens per session scan A 8f634b3adef2

Subscribe to this mod's changes

context-hunter is a skill published in the GitHub repository MrLesk/agents-council (66 stars, last pushed 6mo ago), licensed MIT. It adds 39 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 2026-08-30.

Related

Other skills, from other repositories