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/goccy/sheena/claude-mdgit clone --depth 1 https://github.com/goccy/sheenaWrote 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/goccy/sheena/claude-md)<a href="https://agentmods.dev/instructions/goccy/sheena/claude-md"><img src="https://agentmods.dev/badge/instructions/goccy/sheena/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 | $0.00900 | $0.00900 |
| Opus 5 | $0.00450 | $0.00450 |
| Sonnet 5 | $0.00180 | $0.00180 |
| Haiku 4.5 | $0.00090 | $0.00090 |
Grade A, and why
sheena 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.
How it starts
The opening of the file, as written. The whole thing — 74 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Sheena — Project Rules for Claude
Sheena is a lightweight, in-process sandbox for safely executing AI-agent tool
calls on the same host as the agent. See DESIGN.md for the full design.
Development rules
These rules are mandatory for every change in this repository.
Go version & idioms
- Target Go 1.25.0 as the base. Any feature available up to and including
Go 1.25 may be used freely (generics,
io/fs.ReadLinkFS, etc.). - Prefer modern idioms. Use
any, never the legacyinterface{}spelling. Use generics where they remove duplication or improve type safety instead of falling back toany+ type assertions. go.moddeclaresgo 1.25.0; keep the golangci-lintrun.goin sync.
Testing & coverage
- Every package MUST maintain ≥90% statement coverage. Add unit tests until the threshold is met; do not lower the bar.
make test(runsgo test -race ./...) andmake cover(per-package ≥90% gate) MUST pass before a change is considered done. Both run in CI (see.github/workflows/ci.yml), alongside a separatemake lintjob.- Access-control behavior MUST be verified with Go example tests
(
Example…with// Output:blocks) — e.g. path traversal blocked, network denied by default, overlay never writes to the host. These both document the API and assert the security guarantees. - Keep dedicated security tests for escape attempts (
../, symlink escape, absolute paths, SSRF to loopback/link-local, redirect crossing, memory bomb, command-count overflow, timeout).
Benchmarks
- Sheena claims to be a lightweight sandbox: many instances, low per-call overhead. That claim MUST stay measured, not asserted.
- Keep
Benchmark…tests for: sandbox construction / mass startup (reportallocs/opandB/op, not just ns/op), per-callExecBashoverhead on a trivial command, and parallel throughput (b.RunParallel). - Run with
make bench(go test -run '^$' -bench . -benchmem ./...). When a change may affect hot paths, compare before/after withbenchstatand do not regress startup time or allocations without a stated reason.
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 · 74 lines · 900 tokens per session scan A e39f1385de14
sheena CLAUDE.md is an instructions file published in the GitHub repository goccy/sheena (22 stars, last pushed 1mo ago), licensed MIT. It adds 900 tokens to every session, about $0.0045 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
microsandbox AGENTS.md
AGENTS.md instructions for superradcompany/microsandbox, covering agents.md, scope, project map, design principles and backward compatibility review.
agentos CLAUDE.md
Claude Code instructions for rivet-dev/agentos, covering agentos, boundaries, security model, sqlite schema ownership and runtime and registry.
ironcurtain CLAUDE.md
Claude Code instructions for provos/ironcurtain, covering claude.md, general workflow, git & worktrees, git workflow and platform considerations.
ironcurtain security-boundary.instructions.md
Instructions for provos/ironcurtain, covering security boundary review rules, mandatory checks and structural invariant order.
ironcurtain types.instructions.md
Instructions for provos/ironcurtain, covering types review rules and mandatory checks.
MultiGen CLAUDE.md
Instructions for lingyuanli/MultiGen, covering claude.md, architecture, development commands, full stack (docker) and api (local).