Skip to content

gooseworks-ai/goose-skills / skills/shared/meta-ad-manager

meta-ad-manager

Unverified

Front door and loop owner of a Meta (Facebook/Instagram) ad harness. Reads the brand's ads/ docs, works out which campaign the user means and which stage it is at, hands off to the stage skill that runs it (intake, strategy, creative, launch, checks, fixes, answers), and writes every stage change back to the docs. Use it when someone wants to run ads on Meta, comes back to ads they started earlier, a stage skill hands back, or a scheduled check fires.

references/direct-meta-adapter.md

View on GitHub
# Direct Meta adapter

## Summary

Use this mode when there is no GooseWorks, but a Meta Marketing API access token and ad account
are available. The `ads/` docs are a local folder; Meta is read and written through the Graph
API directly. Launch goes through `launch-meta-ad-campaign`'s own direct adapter, which creates
everything paused and reads it back.

## The `ads/` docs

- A local `ads/` folder in the working directory the user chose for this brand. Create it on the
  first write. One folder per brand.
- Write `local` for `brand_id` and `coworker_agent_id` in `ads/README.md` and `ads/brand.md`.
- Read and write whole files with the host's file tools. Check the folder from the directory that
  holds `ads/`: `python3 <this skill's folder>/contract/validate.py ads`.

## Credentials

Keep the token in the environment (the launch skill's direct adapter names the variables:
the access token and the ad account id). Never put a token in the docs, a prompt, a command
argument or a log. The token needs `ads_read` for checks and answers, and `ads_management` for
launch and pause.

## Surfaces

- Chat in the agent host. There is no messaging channel: skip any "connect a channel" offer, and
  results reach the user in the chat reply.
- Scheduled checks only if the host has its own scheduler. Without one, run the checks when the
  user asks, and say that nothing checks the ads between conversations. "A line in the report"
  means a line in that check's chat reply.

## Stage → skills and calls

| Stage | What runs |
|---|---|
| intake, strategy | the stage skills, in their direct or planning-only mode. The campaign brief is kept in `strategy.md` (see `ad-management-intake`'s adapter) |
| create, review | a written creative brief per angle (message, proof, format). No generation service; the user supplies or makes the creatives |
| connect | the token and ad account in the environment. If a read fails on auth, say the Meta access needs renewing there. Never ask for the token in chat |
| launch | `launch-meta-ad-campaign`, direct Meta adapter. Record the campaign id it returns in `state.md` (`meta_campaign_id`); leave `meta_push_ids: []`. That campaign id is the **launch record**: every ad under it is the harness's own |
| turn on | the direct writer has no activation command. The user switches the campaign on in Ads Manager. Move `launch → live` (or `paused → live`) only after a read shows the campaign's `effective_status` is `ACTIVE` |
| quick check, deep check | Graph API insights reads (see `answer-ads-questions`' direct adapter), or `meta-ads-analyzer` when installed |
| fix and adjust | **Pause** only: set the ad, ad set or campaign `status` to `PAUSED` with the Graph API, then read it back. No revert. A budget change is a written recommendation the user makes in Ads Manager, or approves for you to make, every time |
| answers | `answer-ads-questions`, direct Meta adapter |

## Evidence pointers

In `decisions.md`, point at what you read: `meta:<object id> <since>..<until>` (for example
`meta:act_123 2026-09-17..2026-09-23`), or `user request` (see `contract/RULES.md` §1).
rea-changelog-updatemorluto/reaPrepare or rewrite REA release changelogs and GitHub release notes from pinned Git history, with verified contributor thanks and Release Please synchronization.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
academy-guideanthropics/skillsStop and check this skill before finishing any reply to a question about how to use Claude or a Claude product — it recommends matching courses, tutorials, and use cases from Claude Academy (academy.claude.com), Anthropic's learning hub. Trigger on: "how do I", "how can I", "getting started with", "what can Claude do", "teach me", "learn to use"; questions about artifacts, projects, skills, plugins, connectors, MCP; requests about rolling Claude out to a team, class, or organization; and any ask for training materials, onboarding content, or learning resources. Use it when the user is learning how to use a feature or product — not when they are mid-task and just want the task done. This skill composes with other skills: after consulting product documentation to answer how a Claude feature works, also check here for a matching course or tutorial — a docs-grounded answer and an Academy recommendation belong together. Only recommend on a strong match; never invent Academy content.Stars: 198kUpdated Claude CodeCodexVerifiedInstall
claude-apianthropics/skillsReference for the Claude API / Anthropic SDK — model ids, pricing, params, streaming, tool use, MCP, agents, caching, token counting, model migration. TRIGGER — read BEFORE opening the target file; don't skip because it "looks like a one-liner" — whenever: the prompt names Claude/Anthropic in any form (Claude, Anthropic, Fable, Opus, Sonnet, Haiku, `anthropic`, `@anthropic-ai`, `claude-*`, `us.anthropic.*`, `[1m]`); the user asks about an LLM (pricing/model choice/limits/caching) — never answer from memory; OR the task is LLM-shaped with provider unstated (agent/MCP/tool-definition/multi-agent/RAG/LLM-judge/computer-use; generate/summarize/extract/classify/rewrite/converse over NL; debugging refusals/cutoffs/streaming/tool-calls/tokens). SKIP only when another provider is being worked on (overrides all triggers): OpenAI/GPT/Gemini/Llama/Mistral/Cohere/Ollama named in the query; OR `grep -rE 'openai|langchain_openai|google.generativeai|genai|mistralai|cohere|ollama'` over the project hits (run this grep FIRST if no provider named — don't Read the file).Stars: 198kUpdated Claude CodeCodexUnverifiedInstall
doc-coauthoringanthropics/skillsGuide users through a structured workflow for co-authoring documentation. Use when user wants to write documentation, proposals, technical specs, decision docs, or similar structured content. This workflow helps users efficiently transfer context, refine content through iteration, and verify the doc works for readers. Trigger when user mentions writing docs, creating proposals, drafting specs, or similar documentation tasks.Stars: 198kUpdated Claude CodeCodexVerifiedInstall