Diagnose Before Plan
Use this skill for problem or failure inputs that need diagnosis before planning.
Prime Agent Setup
Read ../../../../references/prime-planning-workflow.md and state-and-config.md — both normative — and follow their setup, state, language, and reviewer/criticizer rules. State lives in .git/prime_plans/ via ../../../../scripts/prime_plans_state.py init --workdir <target-workdir>; spawn reviewer/criticizer roles with await rlm(...).
Diagnostic Workflow
- Inspect available evidence first: repository files, logs, tests, command output, stack traces, recent diffs, CI output, environment details, and user-provided symptoms.
- Produce an in-message RCA summary before asking whether to plan. Use at most 5 Whys. Stop with
unknown when evidence is insufficient; do not invent a cause.
- Ask one choice prompt in the configured language that includes the RCA summary:
Create the scoped fix plan - recommended when evidence supports a planning path.Stop after RCA - keep the diagnosis only.Other.Auto-complete.
- On opt-out, stop after the summary; do not write
PROBLEM_ANALYSIS.md.
- On opt-in (or
Auto-complete choosing the recommendation), select the smallest fitting planning skill and follow that skill exactly: first-turn planning question, final scope confirmation, PLAN_v1.md, refinement, and the native Prime Agent execution handoff.
Level Selection
create-a-small-plan: clear root cause, obvious fix shape, few local files, limited risk.create-a-plan: multi-file fix, external API or dependency behavior, compatibility concerns, or research needed.create-a-big-plan: cross-system failure, unclear root cause, redesign pressure, or high safety/recoverability risk.
Do not ask the user to choose the level unless the evidence supports two materially different planning depths and the tradeoff cannot be resolved from the repo.
PROBLEM_ANALYSIS.md
After opt-in, create the selected planning run's .git/prime_plans state and public artifact directory, then write PROBLEM_ANALYSIS.md before PLAN_v1.md. Include: original problem; symptoms and reproduction status; evidence inspected; RCA summary and 5 Whys (ending early with unknown when evidence stops); suspected root cause and confidence; planning skill selected and why; language, reviewer, and criticizer settings used; open diagnostic gaps the plan must address. Pass the original problem, RCA summary, evidence, and PROBLEM_ANALYSIS.md into the selected planning skill.
1---2name: diagnose-before-plan3description: Diagnose failures before creating a Prime Agent plan. MUST USE for bugs, CI failures, test failures, regressions, incidents, broken behavior, root cause, RCA, or debug-why requests before deciding whether to plan; preserve language, reviewer, and criticizer settings in `.git/prime_plans/config.json`; exclude ordinary feature planning, direct implementation-only, factual/explanation, trivial command-only, or explicit no-plan requests.4---56# Diagnose Before Plan78Use this skill for problem or failure inputs that need diagnosis before planning.910## Prime Agent Setup1112Read `../../../../references/prime-planning-workflow.md` and `state-and-config.md` — both normative — and follow their setup, state, `language`, and reviewer/criticizer rules. State lives in `.git/prime_plans/` via `../../../../scripts/prime_plans_state.py init --workdir <target-workdir>`; spawn reviewer/criticizer roles with `await rlm(...)`.1314## Diagnostic Workflow15161. Inspect available evidence first: repository files, logs, tests, command output, stack traces, recent diffs, CI output, environment details, and user-provided symptoms.172. Produce an in-message RCA summary before asking whether to plan. Use at most 5 Whys. Stop with `unknown` when evidence is insufficient; do not invent a cause.183. Ask one choice prompt in the configured language that includes the RCA summary:19 1. `Create the scoped fix plan` - recommended when evidence supports a planning path.20 2. `Stop after RCA` - keep the diagnosis only.21 3. `Other`.22 4. `Auto-complete`.234. On opt-out, stop after the summary; do not write `PROBLEM_ANALYSIS.md`.245. On opt-in (or `Auto-complete` choosing the recommendation), select the smallest fitting planning skill and follow that skill exactly: first-turn planning question, final scope confirmation, `PLAN_v1.md`, refinement, and the native Prime Agent execution handoff.2526## Level Selection2728- `create-a-small-plan`: clear root cause, obvious fix shape, few local files, limited risk.29- `create-a-plan`: multi-file fix, external API or dependency behavior, compatibility concerns, or research needed.30- `create-a-big-plan`: cross-system failure, unclear root cause, redesign pressure, or high safety/recoverability risk.3132Do not ask the user to choose the level unless the evidence supports two materially different planning depths and the tradeoff cannot be resolved from the repo.3334## PROBLEM_ANALYSIS.md3536After opt-in, create the selected planning run's `.git/prime_plans` state and public artifact directory, then write `PROBLEM_ANALYSIS.md` before `PLAN_v1.md`. Include: original problem; symptoms and reproduction status; evidence inspected; RCA summary and 5 Whys (ending early with `unknown` when evidence stops); suspected root cause and confidence; planning skill selected and why; language, reviewer, and criticizer settings used; open diagnostic gaps the plan must address. Pass the original problem, RCA summary, evidence, and `PROBLEM_ANALYSIS.md` into the selected planning skill.
Optim-Agent