solution-scoring
Skill13 starsChanged 5 months agoMITView on GitHub
Assumption mapping, ICE/SIF scoring and business viability check for solution hypotheses — the scoring engine of /solutions (step 7). Use for "какое решение выбрать", "ICE scoring", "assumption map", "prioritize solutions", "оцени решения", "риски этого решения".
In this repository23 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- ambiguity-resolverSkill
- challengeSkill
- deep-thinkSkill
- design-critique-templateSkill
- experiment-designSkill
- ingestSkill
- interview-analysisSkill
- multi-source-signal-synthesiserSkill
- next-advisorSkill
- pipeline-stepsSkill
- post-launch-reviewSkill
- problem-structuringSkill
- product-requirements-docSkill
- setup-initiativeSkill
- strategic-narrative-generatorSkill
- system-design-docSkill
- technical-spec-documentSkill
- tracking-and-funnelsSkill
- ui-pattern-librarySkill
- user-persona-builderSkill
- user-story-generatorSkill
- user-testingSkill
The file
Files
Solution Scoring — assumption map + ICE + viability
The scoring engine of /solutions: turns "we could do A, B or C" into a
ranked, risk-aware bet. Input: ✅/🎯 problems from the registry
(hypotheses.py show <dir>), never from prose.
1. Assumption map (per solution)
List what must be TRUE for the solution to work; classify each:
| Assumption | Type | Risk | Cheapest validation |
| users will notice the entry point | desirability | high | concept test (step 8.5) |
| sellers can fulfil in 48h | feasibility | med | ops interview |
| unit economics survive the discount | viability | high | spreadsheet + finance review |
Rule: the riskiest assumption gets validated first and cheapest —
that's what experiment-design takes as input. A solution whose riskiest
assumption is untestable is a faith project; say so.
2. Scoring — ICE by default, SIF if the PM's profile says so
- Impact: expected movement of the initiative metric — anchor in registry numbers (segment sizes, measured effects), not vibes.
- Confidence — numeric, staged by validation depth (REAL-backed problems justify confidence in the PROBLEM, not the solution): 0.1–0.3 untested idea or external analogy · 0.4–0.6 after a concept test or a competitive analogue with the same mechanism on a comparable audience · 0.7+ only after an experiment on OUR users. A number outside its stage is the same violation as evidence-typing ranges.
- Ease: S/M/L from the dev lead when available, gut S/M/L otherwise (mark INFERRED).
Comparative table, top-1 recommendation with one-paragraph reasoning. Honesty rule: don't inflate Confidence without evidence — the anti-generic self-check applies.
3. Business viability (quick pass, before design)
Unit economics sketch · cannibalization risk · dependencies (teams/systems
→ candidates for status.json dependencies[]) · compliance flags · effort
S/M/L. Verdict per solution:
- RED — any single disqualifier: unit economics negative at target scale, hard compliance blocker, or effort L with confidence ≤0.3. One red parks the solution regardless of its ICE score (a great score on a non-viable solution is how teams burn quarters).
- YELLOW — proceed with a NAMED risk and an owner for it.
- GREEN — no known disqualifiers.
4. Output
output/solution-hypotheses.md per output-formats.md (S-ids, Formula,
win criteria) + link every survivor to its problem:
hypotheses.py set <problem-id> --link-solution S<N> — the OST and the
Solution phase of the coverage map are built from these links. Parked
solutions stay in the file with their red reason — they will be asked
about at the gate.
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
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.

