agentleFS
tang-vutang-vu/social-posting-skills/skills/social-campaign/SKILL.md

social-campaign

Skill9 starsChanged 2 months agoMITView on GitHub

Turn one source artifact (release, README, post, changelog) into a reviewable, platform-native distribution campaign under human control

4 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

3 rules. 0 are common, 3 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 4 edited versions we read keep these.

  • 3

    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:

    Do NOT shorten the same paragraph 12 times — that fails review.

    line 58
  2. Distinctive:

    No mass DMs, fake engagement, vote manipulation, astroturfing, follow/unfollow farming, CAPTCHA bypass, anti-detection, account rotation, …

    line 85
  3. Distinctive:

    Never print or store credentials, cookies, or browser profile data.

    line 92

“Common” is measured: kept by at least 6 in 10 of the 4 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

Social Campaign

Turn ONE source artifact into a safe, reviewable, platform-native distribution campaign. You are the content intelligence; the social-posting-skills CLI is the deterministic plumbing (validation, approval, receipts, drafts, publishing modes). The human approves before anything is published.

Pipeline

Source → Campaign intent → Content graph → Platform adaptations
→ Preview → Human approval → Publish or export draft → Observe → Learnings

Commands (use the CLI; do not hand-roll files)

npx social-posting-skills campaign create --source <artifact> [--preset oss-launch]
npx social-posting-skills campaign graph   <id>          # see the content graph
npx social-posting-skills campaign set-node <id> --id <node> --value "<text>"
npx social-posting-skills campaign adapt   <id>          # regenerate scaffolds
npx social-posting-skills campaign preview <id>          # THE approval package
npx social-posting-skills campaign approve <id> --items a,b | --all
npx social-posting-skills campaign publish <id> [--items a,b] [--mode ...]
npx social-posting-skills campaign export  <id>          # draft packages
npx social-posting-skills campaign record  <id> --item <i> (--url|--failed|--unknown|--partial)
npx social-posting-skills campaign observe <id> --url <u> --views n --likes n
npx social-posting-skills campaign learn   <id> --text "<lesson>" [--inferred]

Or the same operations as MCP tools (social-posting-skills mcp).

Your job at each step

  1. Source — accept README, CHANGELOG, release notes, blog post, URL, owner/repo@tag, npm:pkg, git:<range>, or pasted text. Never fetch private/authenticated URLs; the fetcher refuses private hosts anyway.
  2. Intent — ask if unclear: goal (launch/explain/milestone), audience, key message, disclosures ("I built this", "Open source", "Sponsored"). A campaign is not "post this everywhere" — pick platforms deliberately.
  3. Content graph — THIS IS WHERE QUALITY LIVES. The CLI extracts a scaffold graph (core-claim, context, proof, technical, demo, cta, links). Refine every extracted node: sharpen the claim, pick the 3 best proofs, keep ONE fact layer — every platform renders from this same graph, so fix facts once, not twelve times.
  4. Adaptation — campaign adapt produces structural scaffolds per platform. Rewrite each into genuinely native content using the platform skill (post-x, post-reddit, ...). Update via: campaign set <id> --item <platform> --field body --value "<text>" Do NOT shorten the same paragraph 12 times — that fails review.
  5. Preview — always show the approval package. It lists per item: platform, account, full content, links, media+alt text, thread shape, warnings, char usage, disclosures, publish method.
  6. Approval — the human picks items. Items with validation errors cannot be approved; fix them first.
  7. Publish / export — three modes, chosen per item:
    • manual → complete draft package (files + instructions)
    • api → official API if the adapter implements it and env creds exist
    • agent-browser → a playbook.json you execute with your own browser tools. Follow it literally: bounded waits, exact selectors, and abort on ambiguity or any challenge/CAPTCHA — never click a look-alike, never bypass a challenge.
  8. Outcome — after browser publishing, record what happened: published (with URL), partial (say exactly which posts landed), failed, or unknown (ambiguous → do NOT re-submit; reconcile first).
  9. Observe & learn — import metrics later (campaign observe). Store lessons with honest labels: --text for observed facts, --inferred for hypotheses. Never claim a pattern "caused" reach.

Hard rules (non-negotiable)

  • No mass DMs, fake engagement, vote manipulation, astroturfing, follow/unfollow farming, CAPTCHA bypass, anti-detection, account rotation, or rate-limit circumvention. Ever.
  • Approval before publishing is the default. --allow-unapproved exists only for explicitly trusted, configured automation — do not default to it.
  • Duplicate public posts are worse than failures. Ambiguous state → record unknown, give reconciliation steps, stop.
  • Never print or store credentials, cookies, or browser profile data.
  • Facts come from the source/factBase. If you can't verify a number, cut 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.