agentleFS
vcxcviivcxcvii/rainmaker/skills/pick-my-battles/SKILL.md

pick-my-battles

Skill0 starsChanged 4 months agoMITView on GitHub

Turn account intelligence into content clusters across seven signal types, not just pain points, with a balance check and a completeness gate that refuses a fourth simultaneous cluster while any existing one is under 40 percent covered.

The file

We split this skill into the rules it gives an agent

8 rules. About 0 are common, 2 are distinctive, and 6 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.

  • 6

    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 23
  2. Very specific:

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

    line 31
  3. Very specific:

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

    line 33
  4. Very specific:

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

    line 39
  5. Very specific:

    Never refuses outright. With no strategy, clusters cannot cite a pain point and this skill should not run.

    line 56
  6. Distinctive:

    No cluster without a pain point id or a named signal.

    line 83
  7. Very specific:

    Completeness gate: Before proposing a new cluster, check the completeness of every existing cluster per spec/site-blueprint.md section 8.

    line 84
  8. Distinctive:

    Every cluster maps to a pain point or a named signal, has an intent and a gap classification, the balance check ran, and the completeness …

    line 102

≈ “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

pick-my-battles

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
data/strategy.jsonpain points, personas, competitors as cluster signalsrun know-my-buyer first
data/snapshots/<latest>/gsc.jsonhead queries and existing coveragecontinue, clusters get impressions: 0
data/competitors.jsoncompetitor-led and gap signalsrun beat-my-competitors first for full coverage
data/snapshots/<latest>/crawl.jsonexisting URLs, so gaps are measured against realityrun rainmaker audit

Produces

clusters (owner). target_tier is shared with follow-the-money.

Refuses when

Never refuses outright. With no strategy, clusters cannot cite a pain point and this skill should not run; direct the user to know-my-buyer first.

Opportunity typology

Clusters come from seven signal types. A system that derives clusters only from pain points misses the fastest commercial content there is.

TypeSignal sourceTypical format
Competitor-lednamed in interviews or in competitors.jsonalternatives, comparison
Objection-ledobjections in business.mdguide, proof, pricing explainer
Feature-ledcapabilities that gate plans or recur in salesfeature page, integration page
Use-case-ledjobs customers hire the product foruse-case page
Vertical-ledindustries named in the ICPvertical landing page
Pain-ledbuyer_language before they knew the category existedsolution article
Winner expansionpages already ranking, from gsc.jsonadjacent page in the same cluster

Procedure

  1. One cluster per validated pain point, minimum. Pain points marked hypothesis produce clusters marked gap: 'missing', ordered below validated ones.
  2. Work through the other six signal types against strategy.json, competitors.json and context/business.md. Every cluster must trace to at least one pain point id or one named signal; a cluster that exists only because a keyword has volume is exactly what this skill exists to prevent.
  3. For each cluster: choose a head query from GSC where an existing query matches the pain or signal language; otherwise propose one and mark impressions: 0.
  4. Attach existing URLs from crawl.json by matching normalised path and title tokens. Set gap: none when a tier-appropriate page exists, thin when it exists under 600 words or without schema, missing otherwise.
  5. Set intent from the query-intent table in skills/_shared/revenue-tiers.md.
  6. Balance check: if any one signal type is more than half of all clusters, name the imbalance explicitly. Competitor-led content alone does not build topical authority; it trades on authority the site does not yet have.

Decision rules

  • No cluster without a pain point id or a named signal.
  • Completeness gate. Before proposing a new cluster, check the completeness of every existing cluster per spec/site-blueprint.md section 8. Refuse to open a fourth simultaneous cluster while any existing one sits below 40 percent covered. Three half-covered clusters beat nothing; six quarter-covered clusters beat nothing at all.
  • Where completeness cannot be computed because no SERP capture has run, say so; the gate reports "cannot evaluate" rather than silently passing or blocking.

Output

## Clusters

<cluster id>: <head query>, intent <x>, target tier <n>, gap <none|thin|missing>
  Signal: <type>, <specific evidence>
  Existing: <url or "none">

Balance: <type> is <n>% of clusters (<flag if over 50%>)
Completeness gate: <passed / refused a new cluster, naming which existing cluster is under 40%>

Done when

Every cluster maps to a pain point or a named signal, has an intent and a gap classification, the balance check ran, and the completeness gate either passed or named the blocking cluster.

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.