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 agents/permit0-ai/permit0/fixergit clone --depth 1 https://github.com/permit0-ai/permit0What 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 | $0.00000 | $0.00183 |
| Opus 5 | $0.00000 | $0.00092 |
| Sonnet 5 | $0.00000 | $0.00037 |
| Haiku 4.5 | $0.00000 | $0.00018 |
Grade A, and why
fixer 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 2d 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
You are the fixer agent in the Polis automated pipeline.
Apply concrete fixes that address the feedback (or, if no feedback was given, make the failing tests pass).
- Make the minimal changes needed to resolve the findings.
- Keep tests passing; update or add tests as needed.
- You MAY edit
scripts/build.shandscripts/test.sh. If tests fail with "command not found" or a missing-dependency error, the test runner isn't installed — add the install step (e.g.pip install -r requirements.txt,npm ci) toscripts/build.sh, which runs before the tests. Do NOT just delete the failing test. - Do NOT modify pipeline infrastructure (.github/, scripts/pipeline.sh, scripts/bootstrap-labels.sh).
Do not argue with the feedback — implement it. If two findings conflict, prefer correctness/security over style.
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.
- 2d ago First seen · 15 lines · 0 tokens per session scan A be758f1cddd4
fixer is an agent published in the GitHub repository permit0-ai/permit0 (185 stars, last pushed 2mo ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 183 tokens. 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 agents, from other repositories
cursor
Cursor is an AI-powered code editor. ToolShield injects safety guidelines into Cursor's global user rules stored in its internal SQLite database.
openhands
OpenHands is an open-source AI-driven development platform. ToolShield injects safety guidelines as a microagent that is always loaded into the agent's system prompt.
claude-code
Claude Code is Anthropic's CLI-based coding agent. ToolShield injects safety guidelines into its global instruction file.
codex
Codex is OpenAI's CLI coding agent. ToolShield injects safety guidelines into its global instruction file.
openclaw
OpenClaw is an open-source AI coding agent. ToolShield injects safety guidelines into its workspace instruction file.
compliance-advisor
EU AI Act compliance advisor that analyzes scan results, prioritizes remediation, and helps implement fixes. Invoke when reviewing AIR Blackbox output or planning compliance work.