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/foxfire1st/agents-remember/agents-mdgit clone --depth 1 https://github.com/Foxfire1st/agents-rememberWhat 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 | $0.02557 | $0.02557 |
| Opus 5 | $0.01278 | $0.01278 |
| Sonnet 5 | $0.00511 | $0.00511 |
| Haiku 4.5 | $0.00256 | $0.00256 |
Grade A, and why
agents-remember 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 3d 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 — 208 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agents Remember Source Checkout Instructions
This repository is the source package for Agents Remember. It is not the live coordination runtime after installation.
If this file is being read from a workspace-level pointer while you are working on a sibling repository, use the installed runtime instructions instead:
<workspace>/ar-coordination/AGENTS.md
When working on this repository itself, use agents-remember as the target
code repository for resolver, onboarding, workflow, and closeout commands.
Start Here — Route By Role
Sessions route by role through the l-01-agent-lifecycles skill. A spawned
agent follows its brief. A developer-facing session is free chat: it answers
research inline and, for role-shaped work, creates or resolves a durable sprint
and first leaf before launching that sprint's architect. It is never a global
architect identity.
Classify the job (bug / feature / triage / research) as a lens during
reframe-research — a hint, re-pickable, never a gate.
The build decision is taken at decide. Chat is never a build route — every
code change lives under an approved task doc:
- Research-only exit — answers or assessments that change no code: no worktree, no task file; chat is the right medium.
- Durable task —
w-02-light-task-workflow: a task document with checklist, decision log, and proposed code examples; small code work takes the minimalw-02-light-task-workflowartifact, and the work escalates to a master + light sub-task series when it outgrows a single-page plan.
The task-collaboration doctrine in tasks/AGENTS.md applies inside the
architect lifecycle's reframe-research phase, in plain chat, before any task
file or format is chosen.
IMPORTANT:
Do not change code or documentation without entering the architect lifecycle and clearing its plan gate.
Do not change task plan items without approval. Think before acting.
Do not randomly commit. Use the c-12-closeout skill instead!
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.
- 3d ago First seen · 208 lines · 2,557 tokens per session scan A f40efc45d6e1
agents-remember AGENTS.md is an instructions file published in the GitHub repository Foxfire1st/agents-remember (27 stars, last pushed 3d ago), licensed MIT. It adds 2,557 tokens to every session, about $0.0128 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.
Other instructions, from other repositories
Perenna AGENTS.md
Instructions for scarletkc/Perenna, covering perenna agent guidelines, start with the project contract, environment, implementation boundaries and generated and versioned artifacts.
krimto CLAUDE.md
Instructions for krimto-labs/krimto, covering krimto — contributor and development guidelines, if you are an ai agent, what this project is, code conventions and required verification (after every change).
krimto AGENTS.md
Instructions for krimto-labs/krimto: This repository is a Claude Code plugin and an MCP server. For full contributor and architecture guidelines, read CLAUDE.md.
krimto GEMINI.md
Instructions for krimto-labs/krimto: See CLAUDE.md for contributor guidelines. Krimto installs as an MCP server (see README.md → "Connect your agent"); the Gemini CLI extension manifest is gemini-extension.json.
dakera-mcp CLAUDE.md
Claude Code instructions for Dakera-AI/dakera-mcp, covering dakera-mcp, key commands, architecture and conventions.
kungfu AGENTS.md
AGENTS.md instructions for kungfu-systems/kungfu, covering agents.md, start from the person's objective, rules that always apply, load bounded context before changing the repository and execute and verify through shifu.