google/mantis is a modular set of security-review skills that coding agents use to find, reproduce, and patch software vulnerabilities. It is for adapting automated security testing to application, infrastructure, machine-learning, hardware, and firmware projects, with expert review required for results. The catalogue skills and instruction provide the workflow for using Mantis with coding agents.
Borrowing it
Nothing to install: this file belongs to google/mantis. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/google/mantis/main/AGENTS.mdgit clone --depth 1 https://github.com/google/mantisWrote 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/google/mantis/agents-md)<a href="https://agentmods.dev/instructions/google/mantis/agents-md"><img src="https://agentmods.dev/badge/instructions/google/mantis/agents-md/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/instructions/google/mantis/agents-md"><img src="https://agentmods.dev/badge/instructions/google/mantis/agents-md.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00824 | $0.00824 |
| Opus 5.5 | $0.00330 | $0.00330 |
| Sonnet 5.5 | $0.00165 | $0.00165 |
| Haiku 4.5 | $0.00082 | $0.00082 |
Grade A, and why
mantis AGENTS.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 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.
How it starts
The opening of the file, as written. The whole thing — 74 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Rules
Before committing
Always run pre-commit run --all-files before creating or amending a
commit.
This ensures all markdown files conform to the pinned mdformat 0.7.21 formatter
(--number --wrap 80 with gfm + frontmatter plugins). The hook installs its own
isolated venv — do not run local mdformat directly.
If the hook modifies files, stage the changes and amend your commit.
ADK Invariant Architecture & Deterministic Gates
Security invariants (INV-1 through INV-6) are enforced deterministically by the Google ADK Python runtime, Pydantic schemas, and tool wrappers rather than prompt fences:
- INV-1 & INV-2 (Evidence & Re-attack): Enforced by
sandbox_tools.pyreached-sink sentinel verification andschemas.pyvalidation. Objective third-party re-attack and strict separation-of-duties role separation are orchestrated bymantis-patchto eliminate self-grading confirmation bias. - INV-3 (Regression Tracking): Enforced by finding lineage schemas and snapshot-match rules.
- INV-4 (Target Immutability & Host Boundary): Enforced by
research_tools.pypath bounds checks (strictly read-only on the host target checkout; all mutations constrained toworkspace/or isolated guest sandboxes). - INV-5 (State Resumption & Monotonic Lineage): Enforced by
BudgetControllercheckpointing anddatabase.pymonotonic status preservation across process restarts via--resume <run_id>. Upstream repository synchronization and environment provisioning are managed externally before resuming Mantis. - INV-6 (Fail-Safe Backward Compatibility): Enforced by schema defaults and fail-closed degradation gates.
Skills must remain clean, dense domain guides focused on vulnerability analysis,
attack vectors, and remediation. Do not re-embed bash Git commands, manual file
locking, or ASCII block fences in skill prompts. All invariants are verified by
reference/tests/test_adk_invariants.py.
Reference files
Skills may split on-demand content into references/ subdirectories (e.g.
skills/mantis-calibrate/references/calibration_rules.md). Before committing,
verify that every references/*.md link target in a SKILL.md file exists on
disk and is tracked by git (git ls-files --error-unmatch <path>).
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 Changed · +2 tokens per session d6c46548011d
- 1mo ago Changed · +13 lines · +152 tokens per session a7801d362080
- 1mo ago First seen · 61 lines · 670 tokens per session scan A 1feb3a9510e2
mantis AGENTS.md is an instructions file published in the GitHub repository google/mantis (2,563 stars, last pushed 2d ago), licensed Apache-2.0. It adds 824 tokens to every session, about $0.0033 per session on Opus 5.5. 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 2026-08-30.
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).
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.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
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 copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).
deno copilot-instructions.md
Copilot instructions for denoland/deno, covering deno development guide for github copilot, network access, high level overview, key directories and key files to understand first.