agent-lighthouse AGENTS.md

Contributor instructions for Agent Lighthouse, a tool that scans websites to check whether AI agents can discover, read, cite, and act on their content. The repository includes command-line, web, report, and MCP components.

In plain words
What is it for?
Use them when changing audit checks, scan orchestration, command-line or web reports, the MCP server, or the website generated from audit evidence.
Why use it?
They set the evidence rule for audits: a report may claim only what documented sources support.

Instructions file for CodexOpenCode

Install

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.

agentmods
npx agentmods add instructions/forkpoint/agent-lighthouse/agents-md
Clone the repo
git clone --depth 1 https://github.com/ForkPoint/agent-lighthouse

Made for: Codex, OpenCode.

Per session 1,803 This file is loaded in full into every session.
When invoked 1,803 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.01803 $0.01803
Opus 5 $0.00901 $0.00901
Sonnet 5 $0.00361 $0.00361
Haiku 4.5 $0.00180 $0.00180

Measured 2d ago against content hash 16f85880350b, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

agent-lighthouse AGENTS.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.

AGENTS.md · 163 lines

How it starts

The opening of the file, as written. The whole thing — 163 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Working in this repository

Agent Lighthouse scans a website and reports whether AI agents can discover, read, cite and act on it. This file is the contributor guide for anyone — human or agent — changing the code.

One rule sits above the rest, and everything below is a consequence of it:

An audit may only claim what a source documents. If no vendor documents a consumer, the check does not affect a site's score. Ever.

Layout

Package What it is
packages/core The audit registry, the scan orchestrator, the gatherer layer. Everything else consumes it.
packages/cli npx @forkpoint/agent-lighthouse <url>
packages/report Terminal, HTML, JSON and Markdown renderers
packages/mcp MCP server exposing a scan as a tool
packages/website The Astro site, built from the registry and docs/evidence/ at build time

Audits live in packages/core/src/audits/<category>/<slug>.ts, one of eight categories: access-crawl-control, agent-interfaces, agentic-commerce, answer-readiness, content-extraction, machine-discovery, operability-safety, structured-data. An audit's id is category/slug, capped at 64 characters by packages/core/src/schemas.ts.

Commands

pnpm build              # every package; core must build before check:dossiers
pnpm test               # vitest, from the repo root only
pnpm typecheck          # tsc --noEmit per package
pnpm lint               # oxlint
pnpm check:dossiers     # registry <-> dossier agreement, both directions
pnpm check:requires     # each audit's `requires` matches what its source reads
pnpm changeset          # one per user-visible change

Read the full file on GitHub · 163 lines

Changes

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.

  1. 2d ago First seen · 163 lines · 1,803 tokens per session scan A 16f85880350b

Subscribe to this mod's changes

agent-lighthouse AGENTS.md is an instructions file published in the GitHub repository ForkPoint/agent-lighthouse (16 stars, last pushed 2d ago), licensed Apache-2.0. It adds 1,803 tokens to every session, about $0.0090 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.