ambiguity-resolver
Skill13 starsChanged 5 months agoMITView on GitHub
Structures a brief HANDED DOWN by a stakeholder (boss, exec, another team) into a one-page problem statement with scope boundaries and a named decision owner. Use when the PM says "мне спустили задачу", "I've been asked to look into Y", "руководство хочет X", "CEO попросил разобраться" — a request the PM didn't originate. For the PM's own vague problem, the FIRST LAUNCH drill-down covers it.
In this repository23 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- 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
- solution-scoringSkill
- 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
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
- Distinctive: line 65
Proposed check-in: After [activity], regroup to decide whether to proceed or pivot.
- Very specific: line 90
Write the answers to output/decisions.md as an ambiguity resolution entry:
≈ “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.
Files
Ambiguity Resolver Skill
Purpose
Turn a brief handed down by someone else into a structured problem statement — so the PM can reply to the stakeholder with clarity instead of booking three more meetings. Distinctive value vs the normal drill-down: explicit OUT-of-scope boundaries and a named decision owner, which protect the PM when the request inevitably mutates.
Three-Stage Process
Stage 1: Reframe
- Restate the vague input as 3-5 explicit questions that need answering
- Identify the unstated assumptions hidden in the brief
- Surface the real decision this feeds into (what will someone do differently once this is resolved?)
Stage 2: Scope
- Define what is explicitly IN scope
- Define what is explicitly OUT of scope (equally important)
- Identify the deadline pressure: is this urgent/important, important/not urgent, or unclear?
- Name who owns the final decision and who needs to be consulted
Stage 3: Action
- Define the minimum viable research: 2-3 activities maximum that would give enough signal to move forward with confidence
- Time estimate for each activity
- What each activity would tell you (and what it wouldn't)
- Proposed check-in point: when to regroup before committing to more
Output Format
Problem Brief: [Opportunity Area]
Restated as questions:
- [Question 1]
- [Question 2]
- [Question 3]
Unstated assumptions we should surface:
- [Assumption 1]
- [Assumption 2]
In scope: [Clear boundary] Out of scope: [Clear boundary] Decision owner: [Name/role] Timeline: [Real deadline if known, or "unclear — recommend setting one"]
Minimum viable research:
| Activity | Time required | What it tells us |
| [activity] | [time] | [insight] |
Proposed check-in: After [activity], regroup to decide whether to proceed or pivot.
Pipeline Integration
This skill is a utility — use it at any pipeline step when input is vague or contradictory.
Common Triggers
| Step | Trigger | Output |
| 0. setup-initiative | PM provides vague brief: "we need to do something about retention" | Full 3-stage process -> feeds into CONTEXT.md |
| 1. analyze-cjm | CJM is incomplete or contradictory | Quick reframe -> clarify scope before hypothesis generation |
| 6. validate-problems | Data partially confirms, partially contradicts | Scope stage -> define what's in/out for solution phase |
| 7. solution-hypotheses | Multiple valid directions, PM can't choose | Reframe stage -> surface the real decision criteria |
| Any step | PM says "I'm not sure what we should do here" | Quick mode (below) |
Quick Mode (3-Question Reframe)
For mid-step ambiguity when the full 3-stage process is too heavy:
- What specifically is unclear? (Name the ambiguity — don't accept "everything")
- What would you do if you knew the answer? (Reveals the real decision this feeds)
- What's the smallest thing we could do to get enough signal? (Minimum viable research)
Write the answers to output/decisions.md as an ambiguity resolution entry:
## YYYY-MM-DD — Ambiguity Resolution at Step N
**Ambiguity**: [What was unclear]
**Real decision**: [What knowing the answer enables]
**Resolution**: [What we decided / what we'll do to find out]
**Next step**: [Specific action]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.

