lamport-problem-statement
Skill0 starsChanged 33 days agoMITView on GitHub
Use this skill to apply Leslie Lamport's method whenever the user is in solution-mode without a clear problem statement. Trigger on: \"should I use X or Y?\", \"I'm thinking of using [approach]\", \"we're debating between A and B\", \"going back and forth on whether to\", \"what are the tradeoffs between\", \"which is better\", \"what should the acceptance criteria be\", or any explicit request to apply Lamport's method. Also trigger when someone proposes a specific technology or architecture before stating what problem it solves. The skill contains an intervention framework and output format — use it rather than engaging with solution details directly when no problem statement has been established yet.
In this repository13 other files it gives its agents.
- AGENTS.mdAGENTS.md
- 1on1Skill
- context-reviewSkill
- gmail-inboxSkill
- learning-labSkill
- logseq-db-querySkill
- nix-sandboxSkill
- obsidian-zettelkastenSkill
- peer-reviewSkill
- portkey-statusSkill
- proton-inboxSkill
- teams-extractorSkill
- theory-of-constraintsSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 17
If the user is already deep in solution-mode — proposing an approach, debating tradeoffs, or asking "should I use X or Y?" — pause and …
- Distinctive: line 44
When producing a problem statement, use this structure:
- Distinctive: line 57
If you're mid-conversation and only have problem + conditions so far, output just those and offer to return once a solution is proposed.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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
Lamport Problem Statement
People naturally jump to solutions. This skill helps redirect that energy: before evaluating how to solve something, get clear on what you're solving and what correct looks like. Once correctness conditions are explicit, comparing solutions becomes objective rather than intuitive.
When to gently intervene
If the user is already deep in solution-mode — proposing an approach, debating tradeoffs, or asking "should I use X or Y?" — pause and check: has the problem been stated? Have correctness conditions been defined? If not, offer to establish them first. Frame it as making the rest of the conversation easier, not as a correction.
Process
- State the problem — What needs to change, and for whom? State it independently of any solution. If the user hasn't stated it, offer a draft and align via conversation. Keep it to 1–3 sentences.
- Define correctness conditions — What must be true for any solution to be considered correct? These are the criteria a solution will be judged against. Draft them if the user hasn't, and refine until they feel complete and specific. (Also called: success criteria, acceptance criteria.)
- Capture the solution (optional) — Once problem + conditions exist, record the proposed solution. If multiple solutions are on the table, this is where they get named.
- Prove or evaluate (optional) — Show how the solution satisfies (or doesn't satisfy) each correctness condition. If conditions aren't met, surface that explicitly.
Output format
When producing a problem statement, use this structure:
Problem: [1–3 sentence statement of what needs solving, independent of solution]
Correctness conditions:
- [Condition 1]
- [Condition 2]
- ...
Solution: [if available]
Evaluation: [how the solution maps to each condition, if both are present]
If you're mid-conversation and only have problem + conditions so far, output just those and offer to return once a solution is proposed.
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.