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/lx-wnk/agent-context/claude-mdgit clone --depth 1 https://github.com/lx-wnk/Agent-ContextWrote 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/lx-wnk/agent-context/claude-md)<a href="https://agentmods.dev/instructions/lx-wnk/agent-context/claude-md"><img src="https://agentmods.dev/badge/instructions/lx-wnk/agent-context/claude-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 5.1 | $0.01206 | $0.01206 |
| Opus 5 | $0.00603 | $0.00603 |
| Sonnet 5 | $0.00241 | $0.00241 |
| Haiku 4.5 | $0.00121 | $0.00121 |
Grade A, and why
Agent-Context CLAUDE.md scanned grade A with 1 finding 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 6d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
When adding a new **shared** file, wire it into `.prompts/setup-prompt.md` Step 2 (download list + parallel curl block). When adding a new **core template**, update `check_critical_templates()` in `install.sh`. Both are How it starts
The opening of the file, as written. The whole thing — 67 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
Project Overview
Agent-Context is a layered, agent-agnostic context architecture for AI coding agents. All layers (0-3) load at startup via @-includes in AGENTS.md, keeping the always-on baseline at ~150-200 effective instruction lines (enforced by tests/check-token-budget.sh). Detailed reference (skills, memory files, delegation/memory-maintenance procedures) is pulled on-demand based on task keywords.
This repository contains the framework itself: shared context files, templates, setup/update prompts, and documentation. It is not a runtime application — it's a template system installed into other projects via .prompts/setup-prompt.md.
Commands
# Format all files
npm run prettier:fix
# Check formatting (CI-style)
npm run prettier
# Run all tests
npm test
Architecture
File Ownership Model
Four categories of files, critical to understand before editing:
- Shared files (
context/, incl.context/bin/,context/hooks/,context/skills/, andcontext/commands/): Overwritten on every auto-update in target projects. Changes here propagate to all installations. Includes thediscovery-mapskill, thecheck-map-budget.shcap validator, and the Claude Code slash commands incontext/commands/(discover,memory-review,decision-review→.claude/commands/) — thin wrappers that invoke the matching skill/review prompt. - Template files (
templates/): Copied once during setup, never overwritten. These become project-owned files (incl.hooks.conf,budget.conf). - Test files (
tests/): CI tests verifyinginstall.shbehavior, template coverage, token budget, memory-prune, hooks, and an offline install smoke test (check-install-smoke.sh, derived from the setup-prompt download table). Not installed into target projects. - Generated artifacts (
outputs/): Reports produced by skills — review findings, handoffs, audits, analyses. Git-ignored, never installed, never referenced by shared files or tests. Regenerate rather than edit; deleting the directory must never break anything.
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.
- 6d ago First seen · 67 lines · 1,206 tokens per session scan A 2c50ac6ed107
Agent-Context CLAUDE.md is an instructions file published in the GitHub repository lx-wnk/Agent-Context (7 stars, last pushed yesterday), licensed MIT. It adds 1,206 tokens to every session, about $0.0060 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
m_flow AGENTS.md
AGENTS.md instructions for FlowElement-xinliuyuansu/m_flow, covering m-flow — developer & agent reference, 1. repository map, extension points, 2. local development and python backend (requires python 3.10 – 3.13).
Meterless AGENTS.md
Instructions for Meterless/Meterless, covering agent instructions for meterless, routing table, isolation rule, repo-wide rules and non-coding agents.
open-bridge AGENTS.md
AGENTS.md instructions for bks-lab/open-bridge, covering the bridge — agent instructions, required reading, session start detection (automatic), theme and agents.
open-bridge copilot-instructions.md
Copilot instructions for bks-lab/open-bridge, covering the bridge — github copilot, run setup once (fresh clone) and session-start gate.
claude-react-kit CLAUDE.md
Claude Code instructions for wahmoh/claude-react-kit, covering claude.md, commands, replace these with your project's actual commands, architecture — mandatory and component rules.
swarm-357 CLAUDE.md
Instructions for TechTideOhio/swarm-357, covering claude.md, what this repo is, python package (packages/techtide-swarm/), or from pypi and key classes.