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/h5i-dev/h5i/claude-mdgit clone --depth 1 https://github.com/h5i-dev/h5iWrote 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/h5i-dev/h5i/claude-md)<a href="https://agentmods.dev/instructions/h5i-dev/h5i/claude-md"><img src="https://agentmods.dev/badge/instructions/h5i-dev/h5i/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 | $0.00447 | $0.00447 |
| Opus 5 | $0.00224 | $0.00224 |
| Sonnet 5 | $0.00089 | $0.00089 |
| Haiku 4.5 | $0.00045 | $0.00045 |
Grade A, and why
h5i 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 today.
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.
What it actually says
CLAUDE.md
This workspace's debug build is very large
Large enough that it has filled a disk. The dependency graph carries a browser
engine and a patched JavaScript engine, so target/debug runs to tens of
gigabytes where target/release is a fraction of it.
That is a fact about the workspace and it applies to anyone who builds it. What
follows from it depends on the machine, so it is not a repository rule: on a
roomy box a dev build is ordinary, and CI builds one on purpose
(.github/workflows/test.yaml runs on throwaway runners, so do not add
--release there).
If your machine cannot spare the space, this repository ships two opt-in guards that make a release-only checkout stick:
-
scripts/no-debug-guard.sh onwrites a regular file attarget/debug, so cargo stops with "failed to create directory ... File exists" before it compiles anything.offandstatusdo what they say. The file lives undertarget/, which is ignored, so it never leaves your checkout. -
scripts/deny-debug-build.pyis a Claude Code PreToolUse hook that refuses dev-profile cargo commands, target-directory redirection, and attempts to remove the guard. It is a script, not a setting: nothing wires it up until you add it to your own.claude/settings.local.json, which is gitignored.{ "hooks": { "PreToolUse": [ { "matcher": "Bash", "hooks": [ { "type": "command", "command": "python3 \"$CLAUDE_PROJECT_DIR/scripts/deny-debug-build.py\"" } ] } ] } }
With both on, every cargo invocation takes --release, the binary is
./target/release/h5i, and scripts that need it default to that path. Do not
work around a "File exists" failure on target/debug: turn the guard off
deliberately, or build release.
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.
- today First seen · 37 lines · 447 tokens per session scan A e78abb35eac0
h5i CLAUDE.md is an instructions file published in the GitHub repository h5i-dev/h5i (617 stars, last pushed today), licensed Apache-2.0. It adds 447 tokens to every session, about $0.0022 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 2026-09-03.
Other instructions, from other repositories
vm0 CLAUDE.md
Claude Code instructions for vm0-ai/vm0, covering claude code project guidelines, development environment, key assumptions, architecture design principles and yagni (you aren't gonna need it).
vm0 AGENTS.md
AGENTS.md instructions for vm0-ai/vm0, a project described as: Zero, your trustworthy AI teammate for real work.
brood-box CLAUDE.md
Claude Code instructions for stacklok/brood-box, covering brood-box, commands — always use task (taskfile.yaml), architecture — strict ddd (domain-driven design), layers and ddd rules (non-negotiable).
cctop AGENTS.md
AGENTS.md instructions for st0012/cctop, covering agents.md - development guide for cctop, scope and sources, required development rules, driver workflow and github and pr rules.
toh-framework CLAUDE.md
Claude Code instructions for wasintoh/toh-framework, covering claude.md — toh framework (repo development guide), what this is, everyday commands, verification protocol and single source, transformed per ide.
agent-rules child-process.instructions.md
Instructions for lirantal/agent-rules, covering system processes secure coding guidelines, your mission and spawning system processes.