adopt
Use when an existing project that was not built with X2 must be migrated to run under it, with documentation sprawl to dissolve,…
@teklabsdigitalInstall SkillDevelopment
Skill
/problem-solve
Use when the product misbehaves at runtime and the cause is unknown, especially when reproducing needs the human's device, eyes, or account. Do not use for a red test with an evident cause (stay in the implement loop) or for green-but-wrong against the locked prototype (that is
Claude Code
$ npx -y skills add teklabsdigital/x2-method --skill problem-solve --agent claude-code
Codex
$ npx skills add teklabsdigital/x2-method --skill problem-solve -a codex -y
Cursor
$ npx skills add teklabsdigital/x2-method --skill problem-solve -a cursor -y
Gemini CLI
$ npx skills add teklabsdigital/x2-method --skill problem-solve -a gemini-cli -y
GitHub Copilot
$ npx skills add teklabsdigital/x2-method --skill problem-solve -a github-copilot -y
OpenCode
$ npx skills add teklabsdigital/x2-method --skill problem-solve -a opencode -y
Windsurf
$ npx skills add teklabsdigital/x2-method --skill problem-solve -a windsurf -y
SKILL.md
problem-solve.SKILL.mdname: problem-solve
description: Use when the product misbehaves at runtime and the cause is unknown, especially when reproducing needs the human's device, eyes, or account. Do not use for a red test with an evident cause (stay in the implement loop) or for green-but-wrong against the locked prototype (that is an adjudication turn; the fix goes in the decision file).
X2 Problem Solve
Diagnosis, priced in human turns. Every "try this and tell me what you see" is a round trip the human pays for, and the metric counts it. The kernel prevents classes of defects; it does not diagnose novel ones, so when one appears, spend the fewest observation turns that honestly find the cause.
The pipeline
1. **Observe.** Get the observation and visual evidence first (screenshot, recording, exact repro). Find the last working state in git; the diff between working and broken is the search space. 2. **Hypothesize.** Form two to four falsifiable hypotheses, each with the data signature it predicts. Write the expected values down before instrumenting; a hypothesis without a predicted signature is a guess. 3. **Validate with one pass.** Design ONE instrumentation deployment that discriminates between all hypotheses simultaneously: five to ten log points, state transitions only, a filterable prefix, no per-render logging. Deploy instrumentation only, no other changes. One round trip. If the data says "working correctly" and the human says otherwise, the hypotheses are wrong, not the human's eyes; return to step 2. 4. **Confirm.** State the root cause in one sentence tied to measured data. If it does not fit in one sentence, it is not understood yet. 5. **Fix.** A failing regression test first where the defect is testable; then the fix through the normal implement loop; then remove every log point added in step 3. If the root cause is a specification defect, that is green-but-wrong: the fix goes in the decision file, never patched into source.
Anti-patterns
- No code changes before a confirmed cause. No layered speculative fixes; if you stack three
changes, you learn nothing. No trying the opposite of a failed fix without new data.
Human-turn contract
- No gate. The observation round trips are the turns; log each one to the ledger like any other,
and design the instrumentation so there is usually exactly one.
What this skill must NOT produce
- No fixes without a measured root cause, no instrumentation left behind, no commits without
direction.
Read more
name: problem-solve description: Use when the product misbehaves at runtime and the cause is unknown, especially when reproducing needs the human's device, eyes, or account. Do not use for a red test with an evident cause (stay in the implement loop) or for green-but-wrong against the locked prototype (that is an adjudication turn; the fix goes in the decision file).
X2 Problem Solve
Diagnosis, priced in human turns. Every "try this and tell me what you see" is a round trip the human pays for, and the metric counts it. The kernel prevents classes of defects; it does not diagnose novel ones, so when one appears, spend the fewest observation turns that honestly find the cause.
The pipeline
1. **Observe.** Get the observation and visual evidence first (screenshot, recording, exact repro). Find the last working state in git; the diff between working and broken is the search space. 2. **Hypothesize.** Form two to four falsifiable hypotheses, each with the data signature it predicts. Write the expected values down before instrumenting; a hypothesis without a predicted signature is a guess. 3. **Validate with one pass.** Design ONE instrumentation deployment that discriminates between all hypotheses simultaneously: five to ten log points, state transitions only, a filterable prefix, no per-render logging. Deploy instrumentation only, no other changes. One round trip. If the data says "working correctly" and the human says otherwise, the hypotheses are wrong, not the human's eyes; return to step 2. 4. **Confirm.** State the root cause in one sentence tied to measured data. If it does not fit in one sentence, it is not understood yet. 5. **Fix.** A failing regression test first where the defect is testable; then the fix through the normal implement loop; then remove every log point added in step 3. If the root cause is a specification defect, that is green-but-wrong: the fix goes in the decision file, never patched into source.
Anti-patterns
- No code changes before a confirmed cause. No layered speculative fixes; if you stack three
changes, you learn nothing. No trying the opposite of a failed fix without new data.
Human-turn contract
- No gate. The observation round trips are the turns; log each one to the ledger like any other,
and design the instrumentation so there is usually exactly one.
What this skill must NOT produce
- No fixes without a measured root cause, no instrumentation left behind, no commits without
direction.
Stats
15
Stars
1
Forks
Maintained
Maintenance
JavaScript
Language
Last commit
Created
Ships withx2
An agentic software engineering method: AI agents do the building, the build enforces the rules that must always hold, a person decides only at the gates, and the method measures its own cost in human turns.
1513 skills
Stars on GitHub
Momentum
- stars in total
- 15
Other skills on x2.
@teklabsdigitalInstall Skilldecompose
Use after the repo is seeded to run the D-000 five-lens decomposition, record persistent schema decisions, define slice one, and…
@teklabsdigitalInstall Skillderive-tests
Use only after the lock (gate 2b) is on record for the slice portion, to derive the fidelity ledger and acceptance scenarios from…
@teklabsdigitalInstall Skilldesign
Use when D-000 exists and the whole-product prototype does not, to run Claude Design in its own context and produce the prototype…
@teklabsdigitalInstall Skillextract
Use when an X2 project completes (or reaches a major milestone) to run the extraction loop, the quality feedback that folds the…
@teklabsdigitalInstall Skillhelp
Use when someone asks what an X2 skill does, which skill to run next, how the gates work, or where an artifact lives. A quick…
@teklabsdigitalInstall SkillMore skills from other plugins.
Auto-invoked
brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or…
@obraInstall Skillask-matt
Ask which skill or flow fits your situation. A router over the skills in this repo.
@mattpocockInstall SkillAuto-invoked
Questions about problem-solve.
- Which coding agents can use problem-solve?
- problem-solve installs in Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot, OpenCode and Windsurf with the skills CLI, using each agent's flag shown on this page. Flowy's auto-invocation, which fires a skill when your prompt matches, runs in Claude Code only.