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/shep-ai/shep/claude-mdgit clone --depth 1 https://github.com/shep-ai/shepWrote 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/shep-ai/shep/claude-md)<a href="https://agentmods.dev/instructions/shep-ai/shep/claude-md"><img src="https://agentmods.dev/badge/instructions/shep-ai/shep/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 | $0.01419 | $0.01419 |
| Opus 5 | $0.00709 | $0.00709 |
| Sonnet 5 | $0.00284 | $0.00284 |
| Haiku 4.5 | $0.00142 | $0.00142 |
Grade A, and why
shep 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 — 93 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
Guidance for Claude Code working in this repository.
Project
@shepai/cli — Autonomous AI-native SDLC platform — run parallel AI agents in isolated worktrees to automate the development cycle from idea to deploy.
Spec Workflow
All feature work MUST begin with /shep-kit:new-feature. See spec-driven-workflow.
/shep-kit:new-feature → /shep-kit:research → /shep-kit:plan → /shep-kit:implement → /shep-kit:commit-pr
Specs live in specs/NNN-feature-name/. Edit YAML only — Markdown is auto-generated.
Commands
| Command | Purpose |
|---|---|
pnpm build |
Build CLI only (fast, for dev) |
pnpm build:release |
Build CLI + web (CI/packaging) |
pnpm test |
Run all tests |
pnpm test:unit |
Unit tests only |
pnpm test:int |
Integration tests only |
pnpm test:e2e |
Playwright e2e tests |
pnpm lint:fix |
Fix lint issues |
pnpm validate |
Lint + format + typecheck + tsp |
pnpm dev:cli |
Run CLI locally (ts-node) |
pnpm dev:web |
Start Next.js dev server |
Architecture
Clean Architecture — four layers in packages/core/src/ (dependencies point inward):
domain/— Core business logic, no external depsapplication/— Use cases, output port interfacesinfrastructure/— External concerns: DB, agents, servicespresentation/— CLI, TUI, Web UI (src/presentation/)
See clean-architecture.
Mandatory Rules
- MANDATORY — TDD: Write failing tests FIRST (RED → GREEN → REFACTOR). Every plan phase must define explicit TDD cycles. See tdd-guide.
- MANDATORY — TypeSpec-first: Domain models defined in
tsp/. Runpnpm tsp:compileto generatepackages/core/src/domain/generated/output.ts. Never edit generated files. See typespec-guide. - MANDATORY — Agent resolution: No component may hardcode an agent type. All resolution flows through
IAgentExecutorProvider. See AGENTS.md. - MANDATORY — Storybook stories: Every web UI component MUST have a colocated
.stories.tsxfile. Commits without stories will be rejected. - MANDATORY — Spec-driven: All features start with
/shep-kit:new-feature. No implementation without a spec. - MANDATORY — Own every failure: You are the ONLY developer. Every test failure, CI failure, and security scan failure is YOUR responsibility. NEVER use the words "unrelated", "pre-existing", or "not our changes". See integrity rules.
- MANDATORY — Read LESSONS.md: You MUST read
LESSONS.mdat the start of every session before writing any code. It contains hard-won lessons from past mistakes. Ignoring it means repeating failures that have already been solved.
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 · 93 lines · 1,419 tokens per session scan A c8a77b0c6363
shep CLAUDE.md is an instructions file published in the GitHub repository shep-ai/shep (247 stars, last pushed 2d ago), licensed MIT. It adds 1,419 tokens to every session, about $0.0071 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
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.
ai-helpers AGENTS.md
AGENTS.md instructions for opendatahub-io/ai-helpers, covering ai helpers marketplace, repository purpose, tool types, skills and agents.
mda CLAUDE.md
Instructions for sno-ai/mda, covering claude.md — sno-ai/mda, release process for @markdown-ai/cli, to cut a release (any 1.1.x, 1.2.0, 2.0.0 — same flow), release notes — non-negotiable and what's new.
pyobfus CLAUDE.md
Claude Code instructions for zhurong2020/pyobfus, covering pyobfus 开发约定, ⚡ current pending work (cold-start 必读), 🟢 2026-09-04 — core 0.5.21 已发布后的当前焦点, ✅ 2026-09-02 — seo 小版本全渠道发布收口 and ✅ 2026-08-24 — 0.5.17 / mcp 0.3.8 发布与外部渠道收尾.
fast-mcp-telegram CLAUDE.md
Claude Code instructions for leshchenko1979/fast-mcp-telegram, covering fast-mcp-telegram, session corrections and 2026-05-27.
free-ai-gateway AGENTS.md
Instructions for zaber-dev/free-ai-gateway, covering 🤖 free-ai gateway - agentic development guidelines, 🏛️ monorepo architecture & package boundaries, 🛑 strict architectural rules for agents, 💻 essential developer commands and build all packages across monorepo.