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/jcast90/relay/agents-mdgit clone --depth 1 https://github.com/jcast90/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 | $0.03051 | $0.03051 |
| Opus 5 | $0.01525 | $0.01525 |
| Sonnet 5 | $0.00610 | $0.00610 |
| Haiku 4.5 | $0.00305 | $0.00305 |
Grade C, and why
relay AGENTS.md scanned grade C 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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
- Tests use per-test tmp dirs. **Never `rm -rf` outside your own tmp dir.** No test should touch a real `~/.relay/`. How it starts
The opening of the file, as written. The whole thing — 137 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Conventions for coding agents (Claude Code, Codex, Cursor, and anything else) working in Relay. This is the short, agent-targeted companion to CONTRIBUTING.md — read both if you're new.
For the product pitch and user-facing features, see README.md. For the deeper walkthrough, see docs/getting-started.md. This doc is about how to work in the code, not what the product is.
Project overview
Relay turns a sentence, a GitHub issue, or a Linear ticket into a running plan of AI-coded work: classifier → planner → decomposer → scheduler → verification → PR watcher. Sessions run inside the user's normal Claude or Codex CLI; Relay wraps them with an MCP server and writes all state (channels, tickets, decisions, crosslink messages) to ~/.relay/ as JSON/JSONL.
The codebase is multi-language by design: TypeScript orchestrator (the CLI, MCP server, orchestration pipeline) + Rust TUI (tui/, ratatui) + Tauri desktop GUI (gui/, React + Vite frontend with a Rust backend) + a shared Rust crate (crates/harness-data/) that the TUI and GUI both use to read ~/.relay/. The three dashboards never talk to each other — they all read the same files on disk.
Development workflow
Most of the loop is live:
- TS edits take effect immediately.
bin/rly.mjsrunssrc/cli.tsthroughtsx, sorly …reflects your source changes on the next invocation. No rebuild. - After
git pulltouchespnpm-lock.yamlorCargo.lock, runrly rebuild. It runspnpm installfirst automatically (pass--skip-installto skip). - Rust TUI changes:
rly rebuild --tui. - Tauri GUI changes:
rly gui --devfor hot-reload, orrly gui --rebuildto refresh the.appbundle.
Set RELAY_USE_DIST=1 if you want to exercise the pre-built dist/ path (marginally faster startup, but stale until rly rebuild).
Verification before pushing
Run these:
pnpm test && pnpm typecheck && pnpm build
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 · 137 lines · 3,051 tokens per session scan C b921c7882fa9
relay AGENTS.md is an instructions file published in the GitHub repository jcast90/relay (5 stars, last pushed 1mo ago), licensed MIT. It adds 3,051 tokens to every session, about $0.0153 per session on Opus 5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
workflow AGENTS.md
AGENTS.md instructions for vercel/workflow, covering agent instructions, architecture, core components, workflow execution model and development commands.
siyuan AGENTS.md
AGENTS.md instructions for siyuan-note/siyuan, covering agents.md, 1. non-negotiable constraints, do not hand-edit, verification and prohibited operations and 2. project-specific rules.
veritas-kanban AGENTS.md
Instructions for BradGroux/veritas-kanban, covering agents.md — canonical agent instructions for veritas kanban, runtime requirements, repository layout, essential commands and install.
NEEDLE CLAUDE.md
Instructions for jedarden/NEEDLE, covering needle project conventions, msrv, module dependency graph, code style and testing.
a-i--skills 03-development-workflow.instructions.md
Instructions for organvm-iv-taxis/a-i--skills, covering development workflow instructions, no repo-wide build system, common development commands, after adding/modifying skills and validation before committing.
google-health-mcp AGENTS.md
Instructions for davidmosiah/google-health-mcp, covering agent development notes, scope, commands and rules.