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/escoffier-labs/agentpantry/agents-mdgit clone --depth 1 https://github.com/escoffier-labs/agentpantryWrote 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/escoffier-labs/agentpantry/agents-md)<a href="https://agentmods.dev/instructions/escoffier-labs/agentpantry/agents-md"><img src="https://agentmods.dev/badge/instructions/escoffier-labs/agentpantry/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 5.1 | $0.01257 | $0.01257 |
| Opus 5 | $0.00629 | $0.00629 |
| Sonnet 5 | $0.00251 | $0.00251 |
| Haiku 4.5 | $0.00126 | $0.00126 |
Grade A, and why
agentpantry 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 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.
How it starts
The opening of the file, as written. The whole thing — 49 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Repository Guidance
Go 1.25 single-module CLI (agentpantry) that syncs browser cookies and named secrets from a daily-driver machine (source role) to an agent machine (sink role) over an encrypted byte stream. Single binary, role chosen by subcommand. Default branch is master.
Definition of Done
Before reporting ANY change complete, run this and confirm it passes:
./scripts/verify
It runs make build, go vet ./..., and go test ./... (includes test/ integration tests).
Conditional gates not covered by the script:
make windowsif you touched anything near a_windows.go/_other.gopair (cmd/agentpantry/,internal/wincrypto)- For security-relevant changes (parsers, framing, filesystem writes, cookie decoding):
make gosec,make vuln, and the matching fuzz target, e.g.make fuzz PKG=./internal/transport FUZZ=FuzzOpen. Fuzz targets exist ininternal/{transport,surface,wire,vault,wincrypto,policy}.
Report actual command results. If anything fails, quote the failure verbatim and do not claim success. CI additionally runs go test -race ./... on Linux and Windows, golangci-lint, govulncheck, and gosec; do not assume local green covers those.
Code Map
- Entry point:
cmd/agentpantry/main.go. All logic underinternal/:source,sink,transport(handshake, AES-256-GCM framing, replay counters),cookie,vault/cdpvault/ffvault/wincrypto(per-browser decryption),policy,surface(sink outputs),secret/secretsrc,config,doctor,keyfile,service,state,wire. - Cross-package integration tests:
test/. Copyable configs:examples/. Threat model:docs/threat-model.md. make packageis the full release gate (test, vet, gosec, vuln, multi-platform build intodist/).
Rules
- Stating a command, flag, or API behavior? Verify it in the code or Makefile first. If you cannot find it, say so. Never invent it.
- Blocked by sandboxing, missing tool, auth, or network? Stop and report the exact blocker and the command that failed. Do not silently work around it or substitute a weaker check.
- A test, fuzz target, lint rule, or CI gate fails? Fix the code. Never weaken, skip, delete, or
t.Skipit to get green. If you believe the test itself is wrong, say so and wait for confirmation. - Touching
internal/transport/handshake.go? Keep both salt-direction paths working:--stdiomode inverts who issues the handshake salt (sink sends it over TCP, source sends it over stdio). Add or update tests for both. - Changing user-facing flow, surfaces, or doctor checks? Update the README claims to match, and bump
CHANGELOG.mdfor releases. - Staging files? Run
git statusfirst.psk.key,config.toml, the rootagentpantrybinary,dist/,memory/, and.brigade/are gitignored on purpose. Never commit keys, configs, or runtime state, and never usegit add -fon ignored paths. - Pushing? Never. Do not push unless the user explicitly asks, and never with
--no-verifyif a pre-push hook exists.
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.
- 5d ago First seen · 49 lines · 1,257 tokens per session scan A 80e1d7b259d9
agentpantry AGENTS.md is an instructions file published in the GitHub repository escoffier-labs/agentpantry (2 stars, last pushed 5d ago), licensed MIT. It adds 1,257 tokens to every session, about $0.0063 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
Uni-CLI AGENTS.md
AGENTS.md instructions for olo-dot-io/Uni-CLI, covering uni-cli, agent routing rule, always-on writing rule, what you can do and web (183+ sites).
opentabs AGENTS.md
Instructions for opentabs-dev/opentabs, covering project instructions for codex, project overview, architecture, you are connected to this project's mcp server and workflow skills.
neobrowser AGENTS.md
Instructions for pitiflautico/neobrowser, covering agents.md — orientation for ai agents & contributors, build, run, test (rust — primary), how it works (rust module map), conventions and ci and release gates.
scout CLAUDE.md
Claude Code instructions for klarlabs-studio/scout, covering claude.md, repository, what this is, commands and build & verify.
empir3-bridge AGENTS.md
AGENTS.md instructions for empir3hq/empir3-bridge, covering empir3 bridge — agent & contributor guide, what lives here, before you edit, run locally and release / publish.
worbrow AGENTS.md
AGENTS.md instructions for noisystreet/worbrow, covering agents.md, project identity, hard constraints (revert on violation), modifying architecture docs and commit conventions.