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/vigolium/piolium/claude-mdgit clone --depth 1 https://github.com/vigolium/pioliumWrote 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/vigolium/piolium/claude-md)<a href="https://agentmods.dev/instructions/vigolium/piolium/claude-md"><img src="https://agentmods.dev/badge/instructions/vigolium/piolium/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 5.1 | $0.02078 | $0.02078 |
| Opus 5 | $0.01039 | $0.01039 |
| Sonnet 5 | $0.00416 | $0.00416 |
| Haiku 4.5 | $0.00208 | $0.00208 |
Grade B, and why
piolium CLAUDE.md scanned grade B with 1 finding 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 6d 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.
Unrestricted tool accessmediumExcessive agency
A wildcard tool grant or "run any command" leaves no least-privilege boundary at all.
Pi packages execute arbitrary code: extensions run TypeScript, skills can instruct the model to run any shell command, and the bundled audit agents declare `bash`, `write`, and `edit` tools. Treat this package as trusted How it starts
The opening of the file, as written. The whole thing — 93 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
Project at a glance
Piolium is a Pi extension (a plugin for @earendil-works/pi-coding-agent) that ships a multi-phase repository security audit pipeline. It is not a standalone CLI — it's loaded into a pi session, which then exposes the /piolium-* slash commands defined here. The bundled sub-agents (agents/) and skills (skills/) are committed to the repo.
Runtime is Bun ≥1.1.0 (see package.json#engines). All Pi SDK packages are listed as peerDependencies and provided by the host pi process.
Commands
bun install
bun run typecheck # tsc --noEmit
bun run lint # biome check .
bun run format # biome format --write .
bun run test # vitest run (fileParallelism: false — audit-state tests touch real fs)
# Run a single test file or filter by name
bun run test -- test/scheduler.test.ts
bun run test -- -t "phase status strip"
# Dev install into Pi (edits in this checkout land immediately, no reinstall)
pi install ./
# Or load the extension ad-hoc without touching settings:
pi -e ./extensions/piolium/index.ts
Driving the extension once installed: pi -p "/piolium-balanced --fresh" (one-shot) or start pi and type /piolium-status. End-to-end harness check: /piolium-smoke.
Architecture
Extension surface (extensions/piolium/index.ts)
The default export pioliumExtension(pi) is the single entry point Pi calls per session. It:
- Registers session flags (
--plm-dir,--plm-since,--plm-scan-limit,--plm-scan-since, and the--plm-*-retries/--plm-*-backoff*family). Each flag mirrors to aPIOLIUM_*env var viaapplyPioliumProcessFlagEnvso downstream modules read a single source of truth (process.env). - Registers a
piolium-streammessage renderer that pretty-prints sub-agent tool-call/result events nested under the originating phase tag. - Registers each
/piolium-*slash command. Every handler delegates to a mode runner underextensions/piolium/modes/and wires phase-strip UI throughcreatePhaseStripCommandUi. - Registers the bundled
anthropic-vertexprovider (providers/anthropic-vertex.ts), but only when Vertex looks configured —isAnthropicVertexConfigured()gates it on thePIOLIUM_VERTEXoverride or any Google/Vertex env var, so a plain install doesn't advertise a provider it can't authenticate.PIOLIUM_VERTEXis read straight fromprocess.env(no--plm-*flag): flag env mirroring runs during command parsing, which is after this init-time registration, so a flag couldn't take effect here.
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.
- 6d ago First seen · 93 lines · 2,078 tokens per session scan B f28a969cfe3f
piolium CLAUDE.md is an instructions file published in the GitHub repository vigolium/piolium (134 stars, last pushed 1mo ago), licensed MIT. It adds 2,078 tokens to every session, about $0.0104 per session on Opus 5. A static security scan graded it B with 1 finding (unrestricted tool access). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).