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.
git clone --depth 1 https://github.com/guerlain-hitier/claude-wardenWrote 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/plugins/guerlain-hitier/claude-warden/warden-invariants)<a href="https://agentmods.dev/plugins/guerlain-hitier/claude-warden/warden-invariants"><img src="https://agentmods.dev/badge/plugins/guerlain-hitier/claude-warden/warden-invariants/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/plugins/guerlain-hitier/claude-warden/warden-invariants"><img src="https://agentmods.dev/badge/plugins/guerlain-hitier/claude-warden/warden-invariants.svg" alt="Reviewed on agentmods" width="80" height="20"></a>Grade A, and why
warden-invariants 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 9d 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.
What it actually says
{
"name": "warden-invariants",
"description": "Optional companion to warden. Re-injects your hard rules on every prompt so they survive context compaction. Adds a UserPromptSubmit hook that reads one local file and writes nothing.",
"version": "0.2.0",
"author": {
"name": "guerlain-hitier",
"url": "https://github.com/guerlain-hitier"
},
"repository": "https://github.com/guerlain-hitier/claude-warden",
"license": "MIT",
"keywords": ["claude-md", "rules", "hooks", "context"]
}
What it installs
The manifest is a name and a version. 1 hook travel with it, and installing the plugin installs all of them. Each is measured on its own page, and each can be installed alone.
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.
- 9d ago First seen · 13 lines scan A 7eaf610f396a
warden-invariants is a plugin published in the GitHub repository guerlain-hitier/claude-warden (1 stars, last pushed 1mo ago), licensed MIT. Its token cost is not measured: this kind of file is read by the harness, not the model. 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 plugins, from other repositories
tools
Session lifecycle management for Claude Code — orient, checkpoint, close out, and never lose the thread between sessions.
engram
Persistent memory for Claude Code. Auto-archives every session to your Obsidian vault, injects recent work at startup, and adds /recall to search your whole history.
recall
Durable, fully-local project memory for Claude Code. Logs every session to .recall/history.md and condenses it into .recall/context.md with a local Python summarizer (TF-IDF + TextRank) — no API key, no external model, nothing leaves your machine. New sessions resume from the saved context.
memstack
131 battle-tested skills for Claude Code. Session memory, git workflows, database migrations, security scanning, deployments, and more. Auto-trigger commands and project handoffs via markdown-based skills.
memory-kit
Persistent memory for Claude Code agents: a hot cache injected every session and held under three size caps, per-session handoffs, and agent-audited promotion into knowledge articles and rules. Carries the builder's layers too — executor/recon/idea-validator agents, session review, second opinion, agent QA and a…
evolving-claude-md
Turn CLAUDE.md into a living Decisions & Learnings log that stays bounded: a format contract enforced at write time by a PreToolUse lint, four downward pressures (strike-through, merge, graduation, quarterly archive), staleness detection against the tree, and a tree-grounded coverage check for the essentials a file…