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/ceorkm/kratos-cli/claude-mdgit clone --depth 1 https://github.com/ceorkm/kratos-cliWhat 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 | $0.01646 | $0.01646 |
| Opus 5 | $0.00823 | $0.00823 |
| Sonnet 5 | $0.00329 | $0.00329 |
| Haiku 4.5 | $0.00165 | $0.00165 |
Grade B, and why
kratos-cli 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 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.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
`~/.claude/settings.json` instead, covering every project on the machine How it starts
The opening of the file, as written. The whole thing — 117 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
Guidance for working in the kratos-memory (a.k.a. kratos-cli) repository.
What this is
A CLI-first persistent memory store for AI coding agents. Memories are saved to
local SQLite databases with FTS5 full-text search — zero network calls, all
data stays on disk under ~/.kratos/. Published to npm as kratos-memory;
exposed as the kratos / kratos-memory binaries.
The deferred mcp__kratos__* tools available in this environment are a separate
MCP wrapper around the same data store — this repo contains no MCP server
code (the src/memory-server/ name is historical; it's just the DB layer).
Commands
npm run build # tsc → dist/
npm run dev # tsx src/cli/index.ts (run CLI from source)
npm start # node dist/cli/index.js (run built CLI)
npm test # builds, then runs node --test against tests/*.test.mjs
Tests (tests/cli.test.mjs) spawn the built CLI in a sandboxed $HOME, so
always npm run build before testing. There is no lint step.
npm install runs scripts/ensure-better-sqlite3.mjs (postinstall), which
self-heals the better-sqlite3 native binary by rebuilding it if the prebuilt
.node file is missing or fails to load.
Architecture
Entry: bin/kratos-cli → loads native-module guard → imports dist/cli/index.js.
src/cli/index.ts— Commander setup. Every subcommand lazilyimport()s its handler fromsrc/cli/commands/<name>.tsso startup stays fast. Adding a command = register it here + add a file incommands/.src/cli/core.ts—initCLIContext()builds theCLIContext: detects the project, opens the project DB and the global DB.getScopedMemoryDb(ctx, {global})picks which DB a command writes to. PII detector is loaded lazily.src/project-manager.ts— Registry-based project detection. No filesystem marker sniffing: cwd (or its deepest registered ancestor) maps to a project. Metadata lives in~/.kratos/projects/<id>/project.jsonwith aprojects.jsoncache that rebuilds from disk if corrupted. Override cwd withKRATOS_PROJECT_ROOT.src/memory-server/database.ts—MemoryDatabase, the core. SQLite + WAL, amemoriestable mirrored into an FTS5 virtual table (mem_fts) via triggers. Scoped per-project (~/.kratos/projects/<id>/databases/memories.db) or global (~/.kratos/global/memories.db).src/security/pii-detector.ts— pattern + entropy based PII/secret detection and redaction. Backssave(warns) andscan.src/compression/—RuleCompressor(viafactory.createCompressor) shortens text onsave --compressand in auto-capture summaries.src/cli/capture-handler.ts+commands/hooks.ts— opt-in auto-capture.hooks installwrites aPostToolUse/Stophook into.claude/settings.local.jsonthat calls the hiddencapturecommand to buffer session activity into a memory.
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 · 117 lines · 1,646 tokens per session scan B 41df0033d423
kratos-cli CLAUDE.md is an instructions file published in the GitHub repository ceorkm/kratos-cli (10 stars, last pushed 25d ago), licensed MIT. It adds 1,646 tokens to every session, about $0.0082 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
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.
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).
next.js AGENTS.md
Instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.