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.
npx agentmods add instructions/rothschildiuk/context-pack/agents-mdgit clone --depth 1 https://github.com/Rothschildiuk/context-packWrote 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.
[](https://agentmods.dev/instructions/rothschildiuk/context-pack/agents-md)<a href="https://agentmods.dev/instructions/rothschildiuk/context-pack/agents-md"><img src="https://agentmods.dev/badge/instructions/rothschildiuk/context-pack/agents-md.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.01171 | $0.01171 |
| Opus 5 | $0.00585 | $0.00585 |
| Sonnet 5 | $0.00234 | $0.00234 |
| Haiku 4.5 | $0.00117 | $0.00117 |
Grade A, and why
context-pack AGENTS.md 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.
How it starts
The opening of the file, as written. The whole thing — 100 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Briefing
context-pack is designed for researchers, squad bots, and personal assistants that explore GitHub repos and surface the most relevant context for new contributors.
[!IMPORTANT] 🤖 CRITICAL DIRECTIVE FOR AI AGENTS: DO NOT use
tree,ls, or randomlycatfiles to discover this repository. To save your context window, you MUST use thecontext-packMCP server. If you support MCP, immediately initialize the server by running:context-pack --mcp-serverIf you cannot use MCP, you MUST run the following command to get your repository briefing before proceeding:
context-pack --cwd . --format jsonFor any non-trivial task, you SHOULD also generate or refresh the local context artifacts before deeper exploration:
context-pack context refresh --cwd .The expected artifacts are:
.context-pack/PROJECT_CONTEXT.md.context-pack/PROJECT_CONTEXT.json.context-pack/memory.mdIf these artifacts are missing or stale, create them before proceeding. If you discover a durable repo truth while working, update
.context-pack/memory.md.
Key signals
- A fast CLI (
context-pack --cwd <repo>) that inspects files, git history, Docker/Compose, dependency manifests, and different guidance docs to summarize the repo in under a few kilobytes. - Configurable budgets (
--max-bytes,--max-files,--max-depth) and filters (--include,--exclude) let you tailor the context bundle to your prompt length or knowledge cutoff. --changed-onlymode and--no-treelet you focus on active work, while--format jsonmakes the output machine-friendly for pipelines.- Selection remains guidance-first but now also applies language-aware boosts using the top detected repository languages for source and entrypoint files.
- Important files include explicit
whyreasons so both markdown and JSON consumers can see why each file was selected. - Built-in regression tests (see
tests/agent_briefing.rsandtests/markdown_snapshots.rs) keep the briefing quality consistent for both markdown and JSON consumers. - For specialized repositories, apply
docs/AI_AGENT_GUIDE.mdas a stricter execution policy for coding agents.
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.
- 4d ago First seen · 100 lines · 1,171 tokens per session scan A 861357fc0147
context-pack AGENTS.md is an instructions file published in the GitHub repository Rothschildiuk/context-pack (10 stars, last pushed 5mo ago), licensed MIT. It adds 1,171 tokens to every session, about $0.0059 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.
Other instructions, from other repositories
aeon CLAUDE.md
Instructions for aeonfun/aeon, covering aeon, how aeon works, strategy, voice and soul file hierarchy (read in this order).
cognirepo CLAUDE.md
Claude Code instructions for ashlesh-t/cognirepo, covering claude.md, key rules, session start sequence (run in this order), behavioral confirmation rule and personas (cognirepo-402, cognirepo-403).
wayland-core copilot-instructions.md
Copilot instructions for FerroxLabs/wayland-core, covering ijfw rules, output discipline, memory routing, context discipline and cross-audit.
mcp-structured-memory CLAUDE.md
Claude Code instructions for nmeierpolys/mcp-structured-memory, a project described as: Structured Memory MCP Server.
inkwell-memory CLAUDE.md
Instructions for veronchenko/inkwell-memory, covering claude.md — inkwellmemory, layout, multi-tenant mode (inkwellmultitenant=1), conventions and testing.
RNR-Enhanced-Cognee AGENTS.md
AGENTS.md instructions for vincentspereira/RNR-Enhanced-Cognee, covering rnr enhanced cognee implementation for codex, critical requirements, 1. ascii-only output (no unicode encoding), 2. dynamic categories (no hardcoded categories) and 3. standard memory mcp interface.