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/future-beat/relay/claude-mdgit clone --depth 1 https://github.com/future-beat/relayWhat 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 | $0.06298 | $0.06298 |
| Opus 5 | $0.03149 | $0.03149 |
| Sonnet 5 | $0.01260 | $0.01260 |
| Haiku 4.5 | $0.00630 | $0.00630 |
Grade A, and why
relay CLAUDE.md scanned grade A 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 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- Used by: External clients (curl, browser dashboard, `scripts/demo.sh`). How it starts
The opening of the file, as written. The whole thing — 251 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Project
Relay — Remaster
Relay is an AI support-triage agent built as a production service: it receives support tickets over a REST API, works each one autonomously with a hand-written agent loop on the Claude API (customer lookup, classification, grounded doc search, reply or escalation), and streams its reasoning as SSE. This milestone remasters the existing v1 into a portfolio showpiece: production-hardened, with real semantic retrieval and a polished live dashboard.
Core Value: A visitor hitting the live demo sees a credible, safe, observably-real AI agent service — impressive to read and watch, cheap to keep running.
Constraints
- Budget: Live demo must stay cheap — single Fly machine, min_machines_running=0, per-run cost budget retained; Voyage usage is index-once + tiny per-query cost
- Tech stack: Python/FastAPI/SQLite/Claude API retained; no orchestration framework — the visible hand-written loop is a feature
- Deploy: One container, no build step, existing Fly.io + GitHub Actions pipeline keeps working
- Compatibility: SSE event contract and MCP tool surface stay backward compatible where practical; evals must keep passing
Technology Stack
Languages
- Python 3.11+ (requires-python
>=3.11inpyproject.toml) - entire application (src/relay/) - HTML/CSS/JS (inline, no build step) -
src/relay/main.py(DASHBOARD_HTMLconstant, a single-page dashboard with vanillafetch/setIntervalpolling) - Markdown - knowledge base content in
kb/*.md, consumed by keyword search tool - Bash -
scripts/demo.sh(demo/smoke script)
Runtime
- CPython 3.11+ (project floor) / 3.12 (CI, Docker) / 3.14 (local venv)
- ASGI server:
uvicorn[standard]>=0.30(pyproject.toml), started viauvicorn relay.main:app(Dockerfile CMD, README quick start) - pip with
pyproject.toml(PEP 621 style), build backendhatchling - No lockfile present (no
requirements.txt,poetry.lock, oruv.lock) — dependency versions are floor-pinned only (>=)
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 · 251 lines · 6,298 tokens per session scan A 20d68911e6d3
relay CLAUDE.md is an instructions file published in the GitHub repository future-beat/relay (0 stars, last pushed 18d ago), licensed MIT. It adds 6,298 tokens to every session, about $0.0315 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). 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.