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/s0fractal/warrantWrote 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/s0fractal/warrant/claude-code-plugin)<a href="https://agentmods.dev/plugins/s0fractal/warrant/claude-code-plugin"><img src="https://agentmods.dev/badge/plugins/s0fractal/warrant/claude-code-plugin/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/s0fractal/warrant/claude-code-plugin"><img src="https://agentmods.dev/badge/plugins/s0fractal/warrant/claude-code-plugin.svg" alt="Reviewed on agentmods" width="80" height="20"></a>Grade A, and why
warrant 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
{
"name": "warrant",
"displayName": "Warrant decision witness",
"version": "0.2.0",
"description": "Signed, offline-verifiable records of the permission decisions Claude Code makes about your agent's tool calls: auto-mode denials, calls that were asked about and then ran, and checks and consequential commands that ran without a prompt. Recalls recent denials at session start.",
"author": { "name": "s0fractal", "url": "https://github.com/s0fractal" },
"homepage": "https://github.com/s0fractal/warrant/tree/master/integrations/claude-code-plugin",
"repository": "https://github.com/s0fractal/warrant",
"license": "MIT",
"keywords": ["audit", "provenance", "permissions", "signatures", "agents"]
}
What it installs
The manifest is a name and a version. 1 skill, 5 hooks travel with it, and installing the plugin installs all of them — 73 tokens a session between 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.
- today First seen · 12 lines scan A 969e84fe6892
warrant is a plugin published in the GitHub repository s0fractal/warrant (5 stars, last pushed today), 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 2026-10-17.
Other plugins, from other repositories
emem
Shared, verifiable memory for AI agents. Two layers, one trust surface. Earth-scale fact substrate: every patch of ground gets a 64-bit address (cell64, about 9.55 m), and every measurement at that cell is a signed fact keyed by (cell, band, tslot), ed25519-signed by the responder over the blake3 of its canonical CBOR.
emem
Verifiable shared memory for AI agents: resolve a place to one canonical address, recall Ed25519-signed Earth-observation facts, and hand another agent a token it can check offline. Reads need no key; the data is maintained by machines (satellites, sensors, open-data connectors), and agents write their own signed…
emem
Shared, verifiable memory for machines and AI. Encode where the data lives, decode with any AI: resolve emem tokens to signed records and check them offline. No API key to read.
bernstein
Drive Bernstein, the open-source governance layer for AI agents, from Cursor. Monitor tasks, agents, costs, approve plans - all without leaving your editor.
agent-lifecycle-kit marketplace
Plugin marketplace listing 1 plugin: agent-lifecycle-kit.
signet
Cryptographic signing for every AI agent tool call — Ed25519 receipts + hash-chained audit log.