agentleFS
vcxcviivcxcvii/rainmaker/skills/check-before-i-publish/SKILL.md

check-before-i-publish

Skill0 starsChanged 4 months agoMITView on GitHub

Run the 10 blocking preflight gates before a page ships, and the 8 postflight gates on their own verification windows after. Never overrides a failing gate; a fail always names the specific fix.

The file

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 2 are distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 4

    Very specific

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

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

    Run rainmaker context --check first.

    line 21
  2. Very specific:

    Only the data/ files listed in this skill's Consumes table.

    line 29
  3. Very specific:

    If strategy.json.context_hash does not match the current hash of context/business.md, say exactly:

    line 31
  4. Very specific:

    If context/business.md carries confidence: stub, continue, and stamp every output with: "Built on a stub context.

    line 37
  5. Distinctive:

    Never auto-override a failing gate: Every failure states the specific fix from the gate's own reason string.

    line 54
  6. Distinctive:

    Every gate reports pass, fail, or not-yet-due with its due date, and every failure names its specific remedy.

    line 94

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill 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
SKILL.md

check-before-i-publish

Context load

Run rainmaker context --check first. It prints what exists, what is stale, and exits 1 if anything this skill requires is missing.

Then read, in this order:

  1. context/business.md in full. If absent, stop: "No business context. Run rainmaker audit, then the know-my-buyer skill."
  2. context/voice.md if this skill writes prose. If absent, stop and say so.
  3. context/glossary.md if this skill names products, features or competitors.
  4. data/strategy.json if this skill reads or writes strategy.
  5. Only the data/ files listed in this skill's Consumes table. Never crawl or call an API the core already covers.

If strategy.json.context_hash does not match the current hash of context/business.md, say exactly:

"Business context was edited after the strategy was written. Re-run know-my-buyer, or run rainmaker context --sync to accept the prose as authoritative."

Then stop.

If context/business.md carries confidence: stub, continue, and stamp every output with: "Built on a stub context. Nothing in it came from a buyer. Run know-my-buyer to replace it."

Consumes

FileWhyIf missing
the brief and draftpreflight inputask for the slug
data/ledger.jsonlshipped timestamps, for postflight windowsrun rainmaker ledger --rebuild
data/snapshots/<latest>/*current measured values for postflight gatesrun rainmaker audit

Produces

Nothing persisted; a pass/fail report.

Procedure

  1. Preflight, before anything ships. Call src/gates/preflight.ts runPreflight(brief), which runs all 10 gates: pain point provenance, cluster slot, cannibalisation, intent match against the SERP verdict, 3+ internal link sources, schema planned, E-E-A-T signals, extractability, the slop check, and the revenue argument. Print the full pass/fail table, not just the failures; a reader needs to see what already passed to trust what didn't.
  2. Never auto-override a failing gate. Every failure states the specific fix from the gate's own reason string.
  3. Postflight, on a schedule after shipping. Call src/gates/postflight.ts postflightStatus(shippedAt, now, measured) per finding, which reports pass, fail, not_yet_due, or unmeasured per gate against its own window (indexed 14 days, canonical and internal links immediate, CWV 7 days, impressions 28 days, position trend, AI citation and conversion contribution 90 days).
  4. A not_yet_due gate is not a failure; report it as pending with the date it becomes due.

Decision rules

  • No gate is ever skipped or overridden. A user asking to publish anyway gets the fail table and is told this skill does not have an override; the decision to ship over a failing gate belongs to the human, not to a bypass built into the tool.
  • Postflight never reports pass or fail before its window elapses. Judging a metric before it has had time to move manufactures both false confidence and false alarm.

Output

## Preflight: <slug>

pain_point_provenance     PASS
cluster_slot              PASS
cannibalisation           FAIL: an existing URL already targets this intent: /old-page
intent_match              PASS
internal_links            FAIL: only 1 inbound link source identified, needs 3
schema_planned            PASS
eeat_signals               PASS
extractability            PASS
slop_check                PASS
revenue_argument          PASS

2 of 10 gates failing. Not ready to publish.

## Postflight: <finding id>
indexed                   not_yet_due (due 2026-08-15)
canonical_correct         PASS
cwv_not_regressed         PASS
internal_links_live       PASS
impressions_appearing     not_yet_due (due 2026-08-29)
position_trend            not_yet_due (due 2026-10-30)
ai_citation               not_yet_due (due 2026-10-30)
conversion_contribution   not_yet_due (due 2026-10-30)

Done when

Every gate reports pass, fail, or not-yet-due with its due date, and every failure names its specific remedy.

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.