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/killertcell428/aigis/claude-mdgit clone --depth 1 https://github.com/killertcell428/aigisWhat 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 | $0.01668 | $0.01668 |
| Opus 5 | $0.00834 | $0.00834 |
| Sonnet 5 | $0.00334 | $0.00334 |
| Haiku 4.5 | $0.00167 | $0.00167 |
Grade A, and why
aigis CLAUDE.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 3d 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 — 159 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Aigis — Project Guidelines
Zenn記事管理（Zenn CLI + GitHub連携）
Zenn記事・書籍はZenn CLIとGitHub連携で管理する。Playwrightやブラウザ操作での投稿は行わない。
ディレクトリ構成
articles/— Zenn記事（slug形式のファイル名:my-article-slug.md）books/— Zenn書籍content/articles/— 他プラットフォーム向け下書き（Qiita, dev.to等）
記事の作成
npx zenn new:article --slug <slug-name>
slugは英数字・ハイフンで12-50文字。articles/ に自動生成される。
プレビュー
npx zenn preview
http://localhost:8000 でプレビュー確認。
記事の公開
- frontmatterの
published: trueに変更 - git commit & push → Zenn側に自動反映
記事の非公開・下書き
published: falseにしてpush
npm scripts
npm run zenn:preview— プレビューサーバー起動npm run zenn:new:article— 新規記事テンプレート生成npm run zenn:new:book— 新規書籍テンプレート生成npm run zenn:list:articles— 記事一覧表示
Auto-Improvement Loop — Release Note & CHANGELOG Format
IMPORTANT: This section OVERRIDES the "one short user-visible sentence" instruction in the loop's Step 7 and Step 11. Write release notes at the level of detail specified here.
Tag ordering — never tag before merging to master
The release workflow (.github/workflows/release.yml) refuses to publish if the tagged commit is not reachable from origin/master (orphan-tag guard). Always:
- Land the
release: vX.Y.Zcommit on master via PR merge first. - Only then create and push tag
vX.Y.Zagainst the master commit.
Pushing a tag from a feature branch HEAD will fail the release workflow and waste a version number. If you hit a "tag collision" error, do not bump the version and retry from the same orphan commit — that is the pattern that produced the v1.1.1 → v1.1.2 → v1.1.3 cascade of unreachable releases. Investigate why master doesn't have the release commit instead.
CHANGELOG entries (Step 7)
Each entry under ## [Unreleased] must cover every new DetectionPattern or detector added in the cycle. For each one, write:
- **`rule_id`** (score N, input/output filter) — One sentence: what attack it detects and what
a blocked example looks like. Include: attack name, source (paper/org/year), and measured
attack success rate if available.
**Blocked example:**
[concrete example of the input or output that would be flagged]
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.
- 3d ago First seen · 159 lines · 1,668 tokens per session scan A b329dce126a2
aigis CLAUDE.md is an instructions file published in the GitHub repository killertcell428/aigis (54 stars, last pushed 5d ago), licensed Apache-2.0. It adds 1,668 tokens to every session, about $0.0083 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
promptsonar AGENTS.md
Instructions for meghal86/promptsonar, covering promptsonar agent instructions, operating principles, repo layout, common commands and change workflow.
promptsonar CLAUDE.md
Instructions for meghal86/promptsonar, covering claude code instructions, promptsonar, structure, key functions and conventions.
Agent-Security-Regression-Harness AGENTS.md
Instructions for OWASP/Agent-Security-Regression-Harness, covering ai agent guide: owasp agent security regression harness, project overview, architecture, contribution workflow and what needs doing.
Project-Koma AGENTS.md
Instructions for swnotmetal/Project-Koma, covering koma — ai coding agent rules, when to use koma, quick integration, design rules and package overview.
claudini CLAUDE.md
Instructions for romovpa/claudini, covering claudini — developer guide, package manager, running benchmarks, run a preset (all methods, all samples) and override method, sample, seed, budget from cli.
SecureAI-Scan CLAUDE.md
Instructions for akanthed/SecureAI-Scan, covering claude.md, what this is, hard requirements (non-negotiable, apply to every change), 1. zero tolerance for false positives and commands.