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/google/mantis/agents-mdgit clone --depth 1 https://github.com/google/mantisWhat 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 | $0.00670 | $0.00670 |
| Opus 5 | $0.00335 | $0.00335 |
| Sonnet 5 | $0.00134 | $0.00134 |
| Haiku 4.5 | $0.00067 | $0.00067 |
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 yesterday.
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 — 61 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.
Verbatim blocks
Blocks A/B/C/D/E/F/G are wrapped in ``` fences and must stay
character-identical across all skills. Never insert content inside their
fences — add notes after the closing ``` instead.
Reference files
Skills may split on-demand content into references/ subdirectories (e.g.
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>).
Extraction criteria
Extract a block into a references/ file only if ALL three hold:
- Off the fail-closed safety-critical path — the extracted content is not
the sole authority for a security-critical invariant (e.g. the
VERIFIED_SECURE => reattack_status = failed_to_bypassgate). - Restates no invariant — the extracted content does not restate any rule that is also stated in the SKILL.md (no invariant may appear in both a SKILL.md and its reference; if a rule must be referenced, state it once and point to it).
- Clean fail-safe fallback — if the reference cannot be loaded, the skill falls back to safe behavior without the reference (e.g. patch rebasing falls back to fresh patch generation, verified by Block G).
mantis-patch/references/patch_rebasing.md qualifies (pure mechanics, no
invariant restated, fails safe to Phase-1). Do not extract content that restates
a crown-jewel invariant — trim it inline instead.
Structural index spec
The structural index spec lives in one canonical location:
mantis-structural-index/SKILL.md. Two other files reference it — the adapter
blueprint stub (mantis-pipeline-adapter/references/mantis-structural-index.md)
and the adapter Guideline 9 (mantis-pipeline-adapter/SKILL.md). Before
committing changes that touch any of these three files, verify:
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.
- yesterday 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 (781 stars, last pushed 4d ago), licensed Apache-2.0. It adds 670 tokens to every session, about $0.0034 per session on Opus 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
kaos-control AGENTS.md
AGENTS.md instructions for kcsinclair/kaos-control, covering claude.md, what this is, authoritative spec, repository layout and build & run.
kaos-control CLAUDE.md
Claude Code instructions for kcsinclair/kaos-control, a project described as: Self-hosted orchestrator for Claude Code agents: drive ideas through a role-gated SDLC (plan → build → test → release), with per-agent sandboxing, live run streaming, and markdown-on-disk as the source of truth. Single Go binary.
kaos-control GEMINI.md
Gemini CLI instructions for kcsinclair/kaos-control, a project described as: Self-hosted orchestrator for Claude Code agents: drive ideas through a role-gated SDLC (plan → build → test → release), with per-agent sandboxing, live run streaming, and markdown-on-disk as the source of truth. Single Go binary.
antigravity-testing-kit GEMINI.md
Instructions for anhtester/antigravity-testing-kit, covering gemini ai - global automation agent rules, git pull restriction rule, browser rules (mandatory), 🖥️ viewport & mode and 🔄 thứ tự debug bắt buộc (playwright mcp).
toh-framework CLAUDE.md
Instructions for wasintoh/toh-framework, covering claude.md — toh framework (repo development guide), what this is, everyday commands, verification protocol and single source, transformed per ide.
awesome-copilot-id AGENTS.md
Instructions for GulajavaMinistudio/awesome-copilot-id, covering communication, explanation and documentation, markdown formatting, user communication style and workflow & methodology.