ActivityPub AGENTS.md

A project guide for Ghost’s ActivityPub server, which lets independent websites publish to the Fediverse, a network of social platforms that can communicate with one another.

In plain words
What is it for?
Use it when working on the TryGhost/ActivityPub codebase, especially for running linting, formatting, type checks, unit and integration tests, end-to-end tests, or migrations.
Why use it?
It gives coding agents the project’s architecture, boundaries, commands, and testing rules so they can change the code without overlooking important conventions.

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/tryghost/activitypub/agents-md
Clone the repo
git clone --depth 1 https://github.com/TryGhost/ActivityPub

Made for: Codex, OpenCode.

Per session 1,808 This file is loaded in full into every session.
When invoked 1,808 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 $0.01808 $0.01808
Opus 5 $0.00904 $0.00904
Sonnet 5 $0.00362 $0.00362
Haiku 4.5 $0.00181 $0.00181

Measured yesterday against content hash df8b9de5479a, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

ActivityPub 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 yesterday.

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 · 159 lines

How it starts

The opening of the file, as written. The whole thing — 159 lines — stays where its author put it; the contents beside it link to each section on GitHub.

ActivityPub — agent guide

A multitenant ActivityPub server for Ghost, built with Fedify, that publishes independent websites to the Fediverse. See README.md for the human overview and /adr for the Architecture Decision Records that own the rationale behind the patterns below.

Stack (versions in package.json / docker-compose.yml): Node + TypeScript, Hono, Fedify, Awilix (DI), Zod, Knex + MySQL, Vitest (unit/integration), Cucumber + Wiremock (e2e), Biome (lint/format), esbuild. Runs in Docker Compose locally; Google Cloud Run in production.

Commands

Tests run inside a Docker container via pnpm, so extra flags passed to pnpm do not reach the test runner — use the dedicated scripts below.

pnpm lint                       # Biome lint
pnpm fmt                        # Biome format
pnpm test:types                 # type check
pnpm test                       # everything (slow)
pnpm test:unit                  # unit only (fast)
pnpm test:integration           # integration only (slow)
pnpm test:cucumber              # e2e only (slow)
pnpm test:single 'path/to/test' # single unit/integration test (fast)

Run a single e2e test by tagging a Feature or Scenario with @only, then pnpm test:cucumber @only.

pnpm dev            # start app (also runs pending migrations)
pnpm logs           # tail app logs  (pnpm dev && pnpm logs to do both)
pnpm stop           # stop app + service deps
pnpm wipe-db        # wipe the database
pnpm fix            # attempt to repair a broken local environment

Local services: nginx :80:8080, MySQL :3307 (user ghost / pass password / db activitypub), Pub/Sub emulator :8085, GCS emulator :4443. Expose port 80 to the internet with tailscale funnel 80.

Migrations

pnpm migration 'name-of-migration'   # create (no spaces in the name)
pnpm migrate                         # apply pending migrations

Migrations also run automatically on pnpm dev. Rollback is unsupported — migrations are forward-only, so treat applied migrations as immutable.

Read the full file on GitHub · 159 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. yesterday First seen · 159 lines · 1,808 tokens per session scan A df8b9de5479a

Subscribe to this mod's changes

ActivityPub AGENTS.md is an instructions file published in the GitHub repository TryGhost/ActivityPub (237 stars, last pushed 3d ago), licensed MIT. It adds 1,808 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.