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/simke9445/agentlocks/agents-mdgit clone --depth 1 https://github.com/simke9445/agentlocksWhat 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 | $0.01031 | $0.01031 |
| Opus 5 | $0.00515 | $0.00515 |
| Sonnet 5 | $0.00206 | $0.00206 |
| Haiku 4.5 | $0.00103 | $0.00103 |
Grade A, and why
agentlocks 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 2d 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.
Repository instructions for agents
Agentlocks is a standalone advisory locking CLI (Bun/TypeScript dev stack, shipped as one Node bundle — a CLI, not a library). Keep defaults generic: do not add prompt-optimization behavior, command aliases, or repository-specific defaults.
Product maturity policy
Agentlocks is not live yet. Do not add compatibility layers, migration paths, deprecated aliases, or fallback behavior for previous internal layouts, schemas, CLI flags, actor outputs, or report formats. When a contract changes, update the implementation, tests, docs, skill instructions, and wiki in place to the new contract.
Commit policy
Make a commit every time a chunk of logic is implemented and unit tested. Do this whether you are working on a branch or inside a worktree: each passing unit-tested chunk gets its own commit before moving on to the next.
Branch and merge policy
Prefer trunk-based development: commit straight to main. agentlocks exists so multiple agents can
share one worktree without clobbering, so branching for isolation is usually unnecessary. Coordinate
with file locks, run bun run check before each commit, and push small green commits to main (see
Commit policy). No branch means no merge, so no history is ever collapsed or lost.
Branch only when a change must be gated before it reaches main (risky or large work needing CI or
review first), is a throwaway experiment, or comes from an outside contributor.
When you do merge a branch, preserve every commit. Never squash.
git merge --no-ff <branch>keeps every commit and the branch topology. Simplest, always safe.- Or rebase the branch onto
mainand fast-forward for a linear history that still keeps every commit; use this only when each branch commit is green. - Keep
mainbisect-safe: every commit, branch or not, must build and pass on its own (the Commit policy guarantees this, which is what makes preserving all commits safe). - Durable narrative (decisions, rejected approaches, lessons) lives in
CHANGELOG.mdand the task ledger, not commit history. - Delete the branch after merge; its commits already live on
main.
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.
- 2d ago First seen · 100 lines · 1,031 tokens per session scan A 1b43cdca5dbb
agentlocks AGENTS.md is an instructions file published in the GitHub repository simke9445/agentlocks (11 stars, last pushed 9d ago), licensed MIT. It adds 1,031 tokens to every session, about $0.0052 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
mandu AGENTS.md
Instructions for konamgil/mandu, covering mandu project guide, package manager, project structure, agent-native workflow and 배포 (release).
mandu CLAUDE.md
Instructions for konamgil/mandu, covering mandu project guide, package manager, project structure, agent-native workflow and 배포 (release).
fff AGENTS.md
AGENTS.md instructions for dmtrKovalenko/fff, covering to clankers, development commands, building, testing and development tools and code quality.
pi-coding-agent-forge AGENTS.md
Instructions for Firstp1ck/pi-coding-agent-forge, covering repository documentation rules, documentation goal, required documentation layers, readme.md — human guide and technical.md — advanced user reference.
rosetta pr-review-bots.instructions.md
How Copilot and CodeRabbit reviews manifest on a rosetta PR, why each looks different from a gh point of view, and what 'wrapping a PR' actually requires.
agentconfig.org AGENTS.md
Instructions for agentconfig/agentconfig.org, covering agent instructions for agentconfig.org, project overview, target audience, site structure and tech stack.