agentsclimarketplace

Adapto content plan

Skill adaptocms/adapto-cms-agent-skills/plugin/skills/adapto-content-plan

A pack of skills for AI coding agents (Claude Code, Cursor) that lets them operate Adapto CMS end-to-end: scaffold projects, design schemas, seed content, translate, run SEO, audit content, roll back.

Install
npx -y skills add adaptocms/adapto-cms-agent-skills --skill adapto-content-plan

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Turn research into a content slate — propose the top content directions for this cycle, work with you to pick and refine them, and produce per-piece briefs the writer executes. Schema-aware (flags when a pick needs a new content type). Writes a dated cycle plan + ledger rows; no CMS writes.

SKILL.md

5.2 KB, as published. Nobody here has run it

adapto:content-plan

The second pipeline step (content-pipeline.md). It reads the latest research dossier and the brain, proposes the top content directions for the cycle, works with you to pick and shape them, and writes a dated cycle plan of per-piece briefs — the spec the writer turns into drafts. It also writes the ledger rows that track each piece. No CMS writes.

Autonomous-safe (studio.md §4): local, no CMS writes — an autonomous cycle can run this unattended; the cycle parks at the human gate before adapto:content-upload.

When to use

  • "Plan this week's content", "what should we write — pick the slate", "turn the research into a plan".
  • After adapto:content-research has produced a dossier.

When not to use

  • Gathering the research itself → adapto:content-research.
  • Writing the drafts → adapto:content-create.
  • Creating the schema a brief needs → adapto:schema-design / adapto:schema-apply.

Inputs

  • The latest dossier(s) in .adapto/research/ (and their menu of opportunities).
  • The brain (pillars, audience, seo, voice) for fit + voice.
  • The ledger (.adapto/ledger.json) — to avoid re-proposing covered/planned pieces and to size the cycle against the cadence.md target (if set).

Outputs

  • A dated cycle plan .adapto/plans/<YYYY-MM-DD>-cycle.md: the chosen pieces, each as a brief (content-pipeline.md §3 — title, slug, type, target/secondary queries, intent, audience, angle, outline, internal links, AEO questions, meta intent, cornerstone, references), each anchored by its slug.
  • Ledger rows for each chosen piece (status: proposedbriefed; pillar, target_query, cycle).
  • Next step: adapto:content-create — write the briefed pieces. (Or adapto:schema-design first if a piece needs a new content type — the schema loop.)

Planning (propose → pick → brief)

  1. Synthesize the dossier + brain into a ranked top-N directions (default 10) for this cycle — each a one-line title + angle + the query/intent it targets + its pillar. Size to cadence.md if set.
  2. Present them as pickable options (conventions.md §10) and work with the user: keep/drop/reshape, add their own ideas, mark which (if any) are cornerstone (top-tier writing). Don't write full content — this is direction-setting.
  3. Schema check (the loop). For each pick, confirm its type exists: built-in article/page, or a custom collection present in .adapto/schema.json. If a pick needs a new collection (e.g. case studies), flag it and route to adapto:schema-designadapto:schema-apply before that piece can be created/uploaded.
  4. Write briefs — expand each kept direction into a full brief (content-pipeline.md §3) in the dated cycle plan, and add/refresh its ledger row. Then state the slate and point to adapto:content-create.

Preconditions

  • Preflight with the adapto:doctor checks.
  • A .adapto/project/ brain and at least one research dossier (run adapto:content-research first; or plan thin from the brain alone and say coverage is limited).
  • No auth/tenant needed (no CMS writes). adapto CLI >= 0.1.3.

Errors and recovery

  • No dossier → suggest adapto:content-research; or propose from the brain alone and flag it's ungrounded.
  • A pick needs a missing content type → flag it and route to adapto:schema-design; don't brief it as uploadable until the schema exists.
  • The user rejects the whole slate → re-propose from a different angle; never force pieces through.
  • Ledger write conflict (concurrent edits) → re-read the ledger, merge the new rows, don't drop existing pieces.

Forbidden actions

  • Never write to the CMS (mutates: false) — plans + ledger rows are local.
  • Never write full content here — briefs only (full drafts are adapto:content-create).
  • Never brief a piece against a content type that doesn't exist — resolve the schema loop first.
  • Never bundle more than one CMS record into a brief. One brief = one record (one article, one page, or one collection item). A piece with companion collection rows (FAQ entries, zones, sources) gets a separate collection_item brief per row — the draft contract has one body and no field for extra records, so a bundled brief leaves the writer nowhere to put them.
  • Never re-propose pieces the ledger already covers without saying so.
  • Plus the global list in forbidden-actions.md — notably: never delete CMS content (archive instead; conventions §9a).

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.