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/codeready-toolchain/tarsy/claude-mdgit clone --depth 1 https://github.com/codeready-toolchain/tarsyWrote 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/codeready-toolchain/tarsy/claude-md)<a href="https://agentmods.dev/instructions/codeready-toolchain/tarsy/claude-md"><img src="https://agentmods.dev/badge/instructions/codeready-toolchain/tarsy/claude-md.svg" alt="Measured on agentmods" 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 | $0.00807 | $0.00807 |
| Opus 5 | $0.00404 | $0.00404 |
| Sonnet 5 | $0.00161 | $0.00161 |
| Haiku 4.5 | $0.00081 | $0.00081 |
Grade A, and why
tarsy 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 5d 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 — 58 lines — stays where its author put it; the contents beside it link to each section on GitHub.
@AGENTS.md
Shared always-on rules are imported from AGENTS.md. This file is the architecture map.
Project Overview
TARSy (Thoughtful Alert Response System) is an intelligent SRE system that processes alerts through parallel agent chains using MCP servers for multi-stage incident analysis and automated remediation. Hybrid Go + Python architecture: Go orchestrator handles business logic, session management, and real-time streaming; a stateless Python service manages LLM interactions over gRPC.
Architecture
Three-service split
- Go backend (
cmd/tarsy/,pkg/) -- Orchestrator, API server (Echo v5), worker pool, event streaming. Port 8080. - Python LLM service (
llm-service/) -- Stateless gRPC server routing to LLM providers (Gemini, OpenAI, Anthropic, xAI, Vertex AI). Port 50051. - React dashboard (
web/dashboard/) -- React 19 + TypeScript + Vite 7 + MUI 7 SPA. Port 5173.
Key Go packages (pkg/)
| Package | Purpose |
|---|---|
agent/ |
Agent framework: IteratingController, SingleShotController, ScoringController, prompt building, tool execution |
api/ |
HTTP handlers (alerts, sessions, chat, review, memory, scoring, trace, timeline, system) |
config/ |
YAML config loading with registries for agents, chains, MCP servers, providers, skills |
database/ |
Ent client wrapper, migrations, GIN index hooks |
events/ |
Real-time event pub via PostgreSQL LISTEN/NOTIFY + WebSocket |
queue/ |
Worker pool with database-backed job claiming (FOR UPDATE SKIP LOCKED) |
mcp/ |
MCP client factory and health monitoring |
memory/ |
Investigation memory: pgvector embeddings, hybrid retrieval (semantic + keyword + RRF) |
services/ |
Domain services (AlertService, SessionService, ChatService, ScoringService, MemoryService, etc.) |
masking/ |
Data masking (K8s Secret detection + regex patterns) |
Data flow
- Alert received via API -> queued in DB
- Worker claims session (FOR UPDATE SKIP LOCKED) -> executes agent chain
- Each agent iteration: Go builds conversation -> gRPC to Python LLM service -> streams response chunks back
- Tool calls executed by Go via MCP servers -> results appended -> next iteration
- Real-time updates via PostgreSQL LISTEN/NOTIFY -> WebSocket to dashboard
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.
- 5d ago First seen · 58 lines · 807 tokens per session scan A 279097f7e5a0
tarsy CLAUDE.md is an instructions file published in the GitHub repository codeready-toolchain/tarsy (10 stars, last pushed yesterday), licensed Apache-2.0. It adds 807 tokens to every session, about $0.0040 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 2026-08-31.
Other instructions, from other repositories
hindsight CLAUDE.md
Claude Code instructions for vectorize-io/hindsight, covering claude.md, project overview, development commands, local development (api + ui) and start both api server and control plane ui.
codedb AGENTS.md
AGENTS.md instructions for justrach/codedb, covering codedb agent guidelines, what codedb is (and isn't), review guidelines, pre-merge verification and security-sensitive areas.
claude-code-settings copilot-instructions.md
Instructions for feiskyer/claude-code-settings, covering claude.md, environment setup, required dependencies, configuration and skills.
pi CLAUDE.md
Instructions for share-skills/pi, covering 强制, 内部目标（不可写入公开文件）, 跑分规范 and eval 驱动优化规则.
harness-sdk CLAUDE.md
Claude Code instructions for strands-agents/harness-sdk, a project described as: Build an agent harness and control it end-to-end. Open-source SDK for production AI agents in Python & TypeScript - any model, any cloud.
the-agency CLAUDE.md
Claude Code instructions for takoyaro/the-agency, covering the agency — working notes, adding a cast member — checklist, shipping a skill, rules that don't bend and model tiering.