agentleFS
graphed-incgraphed-inc/marketing-skills/skills/ad-spy/app/AGENTS.md

marketing-skills / app

AGENTS.md73 starsChanged 3 days agoMITView on GitHub

This project deploys to Graphed cloud from graphed.yaml. It is Ad Spy: a competitor ad library plus a positioning-gaps heat map. The page at / is a Next.js client component that is a React client app. Keep its behavior. Do not wrap it in a scaffold sidebar. - Never guess manifest fields or CLI flags. Run graphed docs manifest or graphed <command> --help first. - Use graphed --format json when you need to parse output. - Migrations are TypeScript Kysely…

  • Reads credentials

The file

We split this AGENTS.md into the rules it gives an agent

3 rules. About 1 is common, 0 are distinctive, and 2 are very specific to this project.

  • ≈1

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Never guess manifest fields or CLI flags: Run graphed docs manifest or graphed <command> --help first.

    line 9
  2. Very specific:

    Use graphed --format json: when you need to parse output.

    line 11
  3. Common:

    No vendor API keys.

    line 19

≈ “Common” is an estimate: at least 1 in 5 of the 41,581 AGENTS.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
AGENTS.md

AGENTS.md

This project deploys to Graphed cloud from graphed.yaml. It is Ad Spy: a competitor ad library plus a positioning-gaps heat map. The page at / is a Next.js client component that is a React client app. Keep its behavior. Do not wrap it in a scaffold sidebar.

Golden rules

  • Never guess manifest fields or CLI flags. Run graphed docs manifest or graphed <command> --help first.
  • Use graphed --format json when you need to parse output.
  • Migrations are TypeScript Kysely modules, modeled on packages/core/src/migrations/0001_ad_spy.ts. Register a new file in packages/core/src/db/migrate.ts. Never edit a migration that has already been applied. Apply with npm run db:migrate (also the deploy release command).
  • Never read process.env directly in app code. Env goes through packages/core/src/config.ts (envSlice() / coreEnv()).
  • Every vendor call is a Graphed Tool or the Graphed OpenRouter proxy (packages/core/src/tools.ts). No vendor API keys.
  • No chat in the app. Users query the database from their coding agent with node cli.mjs sql "<select …>". "Ask your agent" copies that prompt (packages/core/src/prompt.ts).

Project shape

One npm workspace (packages/*).

  • packages/core (@app/core)
    • sources.ts ad library adapters (Meta, Google, LinkedIn via ScrapeCreators) → one ad shape
    • pull.ts list → new ads → media to object storage → rows → queue analysis
    • analyze.ts Gemini 2.5 Flash watches / reads each creative → typed columns
    • research.ts business URL → voice of customer → persona matrix → tag comments and ads
    • gaps.ts gap scores, cell evidence, ad concepts
    • storage.ts project storage with SigV4 (node:crypto), local folder when env is missing
    • api.ts JSON API + /m/<key> media redirect, cli.ts daily job and terminal commands
  • packages/dashboard (@app/dashboard) — Next.js App Router. app/page.tsx renders components/spy. app/api/[...path]/route.ts and app/m/[...key]/route.ts call handleApi.
  • The worker (analysis queue, research resume, ad mapping) starts inside the Next server (packages/dashboard/boot.ts). The daily-pull cron runs node cli.mjs daily.

UI

Hash routes: #/gaps (home: onboarding, then the heat map), #/all, #/new, #/brand/<id>. Styles in app/ad-spy.css (shadcn tokens, white + orange, light only).

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.