agentleFS
runxhqrunxhq/runx/skills/content-pipeline/SKILL.md

content-pipeline

Skill97 starsChanged 33 days agoApache-2.0View on GitHub

Turn governed source evidence into a citation-bound reader-facing draft, channel package, and provider-neutral publication handoff.

17 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

4 rules. 0 are common, 4 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 11 edited versions we read keep these.

  • 4

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use it for articles, updates, newsletters, explainers, and other channel-bound content whose substantive claims need traceable support.

    line 12
  2. Distinctive:

    Stop before drafting when the source packet is missing or invalid.

    line 61
  3. Distinctive:

    Do not turn operator context, campaign intent, or a target URL into evidence.

    line 62
  4. Distinctive:

    Do not add delivery or publication evidence to a local package.

    line 64

“Common” is measured: kept by at least 6 in 10 of the 11 edited versions of this file we could read. “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
SKILL.md

Content Pipeline

This is the standard preparation lane for researched content. It keeps evidence, writing, packaging, and external delivery distinct so an operator can inspect one concrete artifact and no local success is mistaken for publication.

Use it for articles, updates, newsletters, explainers, and other channel-bound content whose substantive claims need traceable support. The result should do something for the reader—clarify a decision, explain a change, establish trust, or enable a concrete next step—not merely prove that research happened.

Composes

<!-- Generated from the native execution closure; run pnpm core-skills:composes:generate. -->
  • ghostwrite#draft
  • ghostwrite#handoff
  • ghostwrite#package
  • research#research

How the chain works

  1. The caller provides the objective, audience, channel, and governed source packets. Usually those packets come from web-fetch or a provider reader.
  2. research admits the sources, verifies citations, and decides whether the evidence supports a useful deliverable.
  3. ghostwrite drafts only from the admitted claims and digests, applying any supplied voice or operator context without treating it as factual evidence.
  4. The validated draft is packaged deterministically for the declared channel.
  5. A provider-neutral handoff records the target and approval requirement. The lane stops before external delivery.

Local research, drafting, and packaging require no approval. A later provider skill owns the consequential gate, idempotency, send or publish request, acknowledgement, and readback.

Inputs and result

  • objective states what the content should change for its reader.
  • source_packets contain bounded governed evidence.
  • audience, channel, and domain shape the artifact.
  • operator_context and target_entities narrow interpretation but are not source evidence.
  • publication_target and boundary_kind (internal or external) prepare the next lane; they do not authorize it. An external delivery needs the delivering lane's own approval.

The result preserves the citation-validated research packet, the evidence-bound draft, a deterministic channel package with not_sent state, and a provider-neutral handoff. needs_more_evidence, needs_review, and not_worth_publishing are valid terminal states when the topic is unsupported, stale, duplicative, or unhelpful.

Stop conditions

  • Stop before drafting when the source packet is missing or invalid.
  • Do not turn operator context, campaign intent, or a target URL into evidence.
  • Withhold any draft whose material claim cannot be tied to an admitted digest.
  • Do not add delivery or publication evidence to a local package.
  • Route the exact accepted artifact into the relevant provider lane instead of rebuilding it there.

Example

An operator wants a blog post explaining a new governance boundary. The lane can validate official docs and receipt evidence, draft for engineering readers, package the accepted body for the blog, and prepare a CMS handoff. It cannot say the post is live until an approved CMS operation and provider readback prove it.

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.