agent-context-os CLAUDE.md

agent-context-os CLAUDE.md is an instructions file for coding agents from conorbronsdon/agent-context-os. It costs 1,153 tokens per session, scanned A, original, MIT.

Project-specific instructions for an agent-context system. They describe identity and project context files, session commands, memory workflows, and rules for when additional context should be loaded.

In plain words
What is it for?
Use them to start or end sessions, save updates, find relevant context, capture notes, reconcile parallel work, and manage the system's memory.
Why use it?
They help the agent preserve useful project and personal context across sessions without loading every file unnecessarily.

Instructions file

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 instructions/conorbronsdon/agent-context-os/claude-md
Clone the repo
git clone --depth 1 https://github.com/conorbronsdon/agent-context-os

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 agent-context-os CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/conorbronsdon/agent-context-os/claude-md.svg)](https://agentmods.dev/instructions/conorbronsdon/agent-context-os/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/conorbronsdon/agent-context-os/claude-md"><img src="https://agentmods.dev/badge/instructions/conorbronsdon/agent-context-os/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 1,153 This file is loaded in full into every session.
When invoked 1,153 The same file — it is already loaded in full.
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 $0.01153 $0.01153
Opus 5 $0.00576 $0.00576
Sonnet 5 $0.00231 $0.00231
Haiku 4.5 $0.00115 $0.00115

Measured 5d ago against content hash 7bc19b39d1ec, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

agent-context-os CLAUDE.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 5d 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.md · 78 lines

How it starts

The opening of the file, as written. The whole thing — 78 lines — stays where its author put it; the contents beside it link to each section on GitHub.

[Your Name] — Context

Who I Am

→ Read identity/who-i-am.md for bio and background.

Slash Commands (Claude Code)

Command Does
/setup Interactive onboarding — builds identity, project, and state files conversationally
/start Load repository state and get a session briefing
/end Log session, update state and the decision log, propose memory updates, check git
/update Mid-session checkpoint — quick state save
/today Morning heartbeat — briefing, staleness check, deadlines
/clean-ai-writing Load writing/skills/avoid-ai-writing/SKILL.md and apply it
/capture Triage raw notes from inbox/ into the right locations
/find-context Find relevant context files by topic keyword (avoids Claude's built-in /context)
/reconcile Drift detection after parallel sessions
/recover Scan orphaned worktrees and stale branches, offer cleanup
/content-shipped Log a published piece of content
/dream Run an on-demand curator pass over the configured memory dir (default: rot detection)
/dream-apply Walk a curator proposal artifact, accept/reject/edit per item
/migrate-gemini Migrate selected Gemini CLI workflows with dry-run review and parity checks
/mine-gemini-workflows Find repeated Gemini workflows and draft portable skills

Add more commands as you build out skills. See docs/agent-template.md for how.

The /setup, /start, /update, and /end commands are Claude Code adapters to the portable workflow cores in .agents/skills/context-*. Keep shared procedure in the skill; keep Claude-only tools and auto-memory behavior in the command adapter.

When to Load Additional Context

For tasks without a slash command → load ROUTING.md.

Thought-Partner Mode

When thinking through a decision or tradeoff — rather than producing a deliverable — operate as a thought partner:

  • Challenge assumptions. If a premise seems weak or unexamined, say so.
  • Offer the alternative. Present the strongest case for a different path.
  • Ask the uncomfortable question. What hasn't been considered or is being avoided?
  • Be direct about tradeoffs. Don't soften the downsides.
  • Summarize the decision cleanly. What's the choice, what does each path require, and what's the real risk?

Read the full file on GitHub · 78 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. 5d ago First seen · 78 lines · 1,153 tokens per session scan A 7bc19b39d1ec

Subscribe to this mod's changes

agent-context-os CLAUDE.md is an instructions file published in the GitHub repository conorbronsdon/agent-context-os (22 stars, last pushed today), licensed MIT. It adds 1,153 tokens to every session, about $0.0058 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.