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/jmrashed/smspit/claude-mdgit clone --depth 1 https://github.com/jmrashed/SMSPitWhat 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 | $0.02934 | $0.02934 |
| Opus 5 | $0.01467 | $0.01467 |
| Sonnet 5 | $0.00587 | $0.00587 |
| Haiku 4.5 | $0.00293 | $0.00293 |
Grade A, and why
SMSPit 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 2d 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 — 113 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
Guidance for Claude Code (or any AI agent) working in this repository.
Project overview
SMSPit is a self-hosted SMS sandbox (like Mailpit, but for SMS) for local development, testing, and CI/CD — it captures outgoing SMS instead of delivering them, so developers can inspect, search, replay, and debug messages without a real SMS provider. Full product description, feature list, and architecture diagram live in README.md — read it before making product decisions.
Current state (check before assuming otherwise)
Progress tracks checklist.md; check it for the exact day-by-day cutoff, but roughly:
README.md— product description, planned architecture, tech stack, and roadmap (v0.1 → v1.0)checklist.md— a 100-day, day-by-day build checklist derived from the roadmap; the source of truth for build order and completion statusLICENSE— MITCLAUDE.md— this filesms-service/(NestJS) — v0.1 message capture/search REST API, replay endpoint, plus API-key auth guard (Days 11–41); has aDockerfileand its own unit + e2e test suitesauth-service/(Laravel) — users/API-key schema, key generation/list/revoke, and validation middleware (Days 31–35, 47); has its own test suitegateway/(Go) — base HTTP server (chi router), health check, path-based reverse-proxy routing tosms-service(/api/v1/*) andauth-service(/auth/*→/api/*), and API-key auth enforcement on the/api/v1/*route (validates against auth-service, forwards identity viaX-Api-Key-Id/X-Owner-Id/X-Api-Key-Scopesheaders) (Days 37–39)dashboard/(React) — inbox, message detail, search/filter, replay, statistics, real-time WebSocket updates, and API key management UI wired to the REST API (Days 21–25, 42–47)docker-compose.yml— wires all six v0.2 services (gateway,auth-service,sms-service,dashboard, Postgres, Redis) (Days 28, 49);ai-serviceandworkerare not yet added (pending Day 79). Nodocker/podmanis available in this working environment, so compose changes are verified by replicating the same URLs/ports as live host processes rather thandocker compose up-- re-verify for real where Docker existsai-service/,worker/— still empty skeletons with stubREADME.mds marked "Not yet implemented" (Phase 4, Days 66+)- Placeholder structural folders:
proto/,docs/,docker/,deployments/,scripts/
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.
- 2d ago First seen · 113 lines · 2,934 tokens per session scan A a064bfa8f259
SMSPit CLAUDE.md is an instructions file published in the GitHub repository jmrashed/SMSPit (5 stars, last pushed 1mo ago), licensed MIT. It adds 2,934 tokens to every session, about $0.0147 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
ecspresso CLAUDE.md
Instructions for kayac/ecspresso, covering claude.md, project overview, git workflow, build commands and build the binary.
homebutler CLAUDE.md
Instructions for Higangssh/homebutler, covering homebutler, verify before claiming anything works, writing commits, prs, and comments, no tool attribution and commit subjects.
picoaide AGENTS.md
Instructions for picoaide/picoaide, covering claude.md, 项目概述, 构建与运行, 构建命令 and ⚠️ 重要：必须全量编译.
claude-tools-mcp CLAUDE.md
Instructions for mathematic-inc/claude-tools-mcp, covering claude.md, project overview, development commands, building and running tests.
go-service-template-rest AGENTS.md
Instructions for Dankosik/go-service-template-rest, covering agents.md, authority, decision ownership, engineering and validation budget.
go-service-template-rest CLAUDE.md
Instructions for Dankosik/go-service-template-rest, a project described as: Go REST API template and Golang microservice boilerplate, AI-native for coding agents, with OpenAPI, PostgreSQL, sqlc, observability, and CI.