agentleFS
Kikobazz123Kikobazz123/lordgen-competition-demo/.claude/skills/opportunity-score/SKILL.md

opportunity-score

Skill0 starsChanged 3 months agoMITView on GitHub

Score the automation opportunities found in data/research.json and rank them, honestly reflecting the confidence of the underlying evidence. Use after business-research has produced data/research.json.

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Before this skill's output is considered done, verify programmatically:

    line 59
  2. Very specific:

    Every factor_scores object has all 8 factors, each 1–5.

    line 60

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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

Opportunity Score

Purpose

Turn data/research.json's automation_opportunities into a ranked, scored shortlist — not a generic report. Never let a low-evidence opportunity score as if it were well-supported (CLAUDE.md §11, §12).

Input

data/research.json, specifically the automation_opportunities array and each opportunity's confidence and supporting_fact_refs.

Rubric

Score each opportunity on 8 factors, 1–5 each, where 5 is always the favorable end (so "risk" and "implementation complexity" are scored inverted — 5 means low risk / low complexity):

Factor5 means
business_impactWould matter a lot to the business if it worked
automation_potentialWell suited to AI/automation vs. needing human judgment
feasibilityBuildable within this project's architecture (Claude Code + n8n + ClickUp, human approval gate)
implementation_simplicitySimple to build (inverse of complexity)
time_savingsMeaningful time saved if it works
revenue_customer_impactMeaningful revenue or customer-experience effect
data_availabilityMust track the opportunity's confidence in research.json — low research confidence caps this low, regardless of how good the idea sounds
risk_levelLow risk if wrong (inverse of risk)

overall_score = round(average(8 factors) * 20) → 0–100 scale, all factors equally weighted (no factor is allowed to dominate — keeps the rubric auditable).

Tiers: 80–100 High · 60–79 Medium · 40–59 Low · <40 Not recommended.

Rule: data_availability may never be scored higher than the source opportunity's confidence justifies (low confidence → data_availability ≤ 2). This is the mechanism that stops weak research from producing an inflated score.

Required Output

Write data/score.json:

{
  "input_source": "data/research.json",
  "scoring_date": "YYYY-MM-DD",
  "scores": [
    { "opportunity": "", "factor_scores": { "...8 factors...": 1-5 }, "overall_score": 0-100, "tier": "", "reasoning": "", "risks": [""], "research_confidence": "low|medium|high" }
  ],
  "top_opportunity": { "opportunity": "", "overall_score": 0, "tier": "", "reasoning": "" },
  "recommendation": "",
  "limitations": [""],
  "approval_required": false
}

approval_required is false here — scoring is internal reasoning, not a consequential action. It exists in the schema for consistency with CLAUDE.md §12's suggested output fields; the real approval gate is the developer sign-off between build phases.

Validation (must run before reporting results)

Before this skill's output is considered done, verify programmatically:

  • Every factor_scores object has all 8 factors, each 1–5.
  • overall_score matches round(average(factor_scores) * 20) — recompute, don't trust the written value.
  • data_availability is consistent with the source opportunity's confidence (low confidence → data_availability ≤ 2).
  • top_opportunity actually is the max-scoring entry in scores.

Out of Scope

This skill only scores and ranks. It does not write the proposal — that's proposal-generator (Phase 4), which should treat top_opportunity as an input, not re-litigate the ranking.

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.