Put It On Autopilot

put-it-on-autopilot

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

Produces

.github/workflows/weekly.yml, .github/workflows/monthly.yml.

Procedure

1. Determine site shape from the latest crawl and GSC snapshot: URL count, clicks per month, and recent publish rate.
2. Recommend a cadence from spec/agent.md section 5:
   • Under 50 URLs or under 100 clicks/month: no weekly job, monthly audit and report, quarterly strategy review only if drift fires.
   • 50 to 500 URLs: weekly pulse (fetch plus audit --refresh), monthly full audit, report, off-site scan, AI probes.
   • Over 500 URLs or publishing 4+ pages a month: full routine weekly, monthly report and off-site scan, ledger --compact.
3. Never assume the cadence; present the recommendation and let the user confirm or override before writing any workflow.
4. Generate the workflow calling rainmaker routine (weekly) and rainmaker report --window month plus rainmaker ledger --compact (monthly).
5. Verify every secret the workflow needs is documented in the README's credentials section, and print the exact gh secret set commands. Never write a workflow depending on a secret the user has not been told to create.
6. AI citation probes are scheduled monthly, never weekly: they cost money per run and the answers are non-deterministic, so a weekly line would mostly measure sampling variance.
7. Built-in crawl is the workflow default. Include --provider firecrawl or --provider contextdev only after the user explicitly approves that provider and its quota or cost.
8. GitHub Actions does not allow secrets.* directly in a step if:. Map an optional secret into job-level env, then test env.NAME != '' in the condition.

Decision rules

• Never re-check anything faster than its own verification window. A weekly ranking report on a 40-page site reports noise as news.
• Never schedule a workflow whose secrets are undocumented.
• Never activate a paid provider because its secret exists.
• Never emit if: ${{ secrets.NAME != '' }}. Use a job-level environment variable and if: env.NAME != ''.
• The cadence is a recommendation the user confirms, never a default silently applied.

Output

## Recommended cadence

Site shape: <n> URLs, <n> clicks/month, publishing <n> pages/month
Recommendation: <weekly / monthly-only / full weekly>, because <reason>

Confirm, or tell me what to change.

## Workflows written
.github/workflows/weekly.yml
.github/workflows/monthly.yml

## Secrets still needed
gh secret set FIRECRAWL_API_KEY --repo <repo>
gh secret set GOOGLE_APPLICATION_CREDENTIALS --repo <repo>

Done when

The cadence was confirmed by the user before anything was written, both workflow files validate, and every required secret was printed with its exact gh secret set command.