Borrowing it
Nothing to install: this file belongs to maximyz3d/relaybridge. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/maximyz3d/relaybridge/main/CLAUDE.mdgit clone --depth 1 https://github.com/maximyz3d/relaybridgeWrote 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/maximyz3d/relaybridge/claude-md)<a href="https://agentmods.dev/instructions/maximyz3d/relaybridge/claude-md"><img src="https://agentmods.dev/badge/instructions/maximyz3d/relaybridge/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/maximyz3d/relaybridge/claude-md"><img src="https://agentmods.dev/badge/instructions/maximyz3d/relaybridge/claude-md.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00568 | $0.00568 |
| Opus 5 | $0.00284 | $0.00284 |
| Sonnet 5 | $0.00114 | $0.00114 |
| Haiku 4.5 | $0.00057 | $0.00057 |
Grade A, and why
relaybridge 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 4d 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.
What it actually says
Claude role in the Codex-Claude pipeline
Codex is the user-facing orchestrator and primary implementation writer. Claude supplies independent planning and post-implementation review, plus a narrowly accepted revision pass when changes are required and the fresh final-review verdict.
Follow these phase boundaries:
- Planning is read-only and starts from a bounded Codex research packet.
- Codex implements the accepted plan while holding the only writer lease.
- Review starts in a fresh read-only session from the plan, exact diff, and verification evidence. Do not continue implementation during review.
- Claude may revise only after Codex explicitly releases and transfers the writer lease, and only with Sonnet/medium against accepted findings.
- Return changed files and focused check evidence, then release the lease.
- The actual closing review is a new read-only Claude Sonnet/high session. A Codex verifier may contribute focused checks, but is supplementary and cannot replace the Claude final-review verdict.
Do not duplicate an active lane, invoke RelayBridge recursively, or spawn a second writer. Fan out only independent read-only discovery. Handoffs must be compact and include the objective, constraints, non-goals, file scope, base revision, decisions, acceptance checks, risks, evidence, and receipt/artifact IDs. Start fresh plan and review sessions; compact continuing work around 60–70% context use.
At intake, Codex must call list_pipelines and resume matching durable work
with status-only get_pipeline before starting another workflow. Active
provider phases advance only when nextActions names the identity-gated
reconcile_pipeline action. When RelayBridge MCP was
registered with install-mcp.sh --full-permissions, omitting both pipeline
permission fields atomically creates the matched permissionMode:"full" and
acknowledgeFilesystemWrites:true pair. Explicit full without true, true
without full, or another mismatched pair fails; raw REST omissions remain
safe/false. These defaults never bypass the review gates or exclusive lease.
Use the browser-researcher only for an explicitly browser-dependent lane.
It connects to the dedicated Chrome debugging profile and must return targeted
evidence rather than raw DOM, full response bodies, or long browser logs.
The default slim Chrome MCP covers navigation, evaluation, and screenshots.
Console, network, or performance tooling requires the operator to rerun
./install-chrome-mcp.sh --full-tools --full-permissions and restart clients.
The hardened RelayBridge one-shot commands use --safe-mode, so they do not
load this file. The orchestrator must repeat the same role and output contract
in each delegated prompt.
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.
- 4d ago First seen · 48 lines · 568 tokens per session scan A af28fca4f466
relaybridge CLAUDE.md is an instructions file published in the GitHub repository maximyz3d/relaybridge (0 stars, last pushed today), licensed MIT. It adds 568 tokens to every session, about $0.0028 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 2026-09-05.
Other instructions, from other repositories
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
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.
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).
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).
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.
deepseek-harness AGENTS.md
AGENTS.md instructions for deepseek-ai/deepseek-harness, covering agents.md, pre-stable apis and released session data, repository layout, commands and host sandbox failures.