$npx -y skills add avibebuilder/claude-prime --skill give-planUse when the user wants a written, reviewable plan or spec produced before coding starts. Triggers on: mapping out changes without implementing, thinking through risks of upgrades or migrations, evaluating approaches before committing to one, writing specs for team review, phasin
| 1 | ultrathink |
| 2 | |
| 3 | ## Core contract |
| 4 | |
| 5 | Figure out the right approach, write a plan the user can inspect, and **stop**. Planning is not approval to implement — wait for explicit go-ahead before writing any code. |
| 6 | |
| 7 | - Clarify only what would materially change the plan. |
| 8 | - Read the codebase before making claims. Distinguish confirmed facts, inferences, and unknowns — never hide uncertainty. |
| 9 | - Optimize for reviewability over momentum. |
| 10 | |
| 11 | If already in native plan mode, this skill shapes *how* to plan; plan mode provides the workflow structure and must follow this skill guidelines. |
| 12 | |
| 13 | ## Process |
| 14 | |
| 15 | Check conversation context and skip completed steps. |
| 16 | |
| 17 | ### 1. Clarify the planning goal |
| 18 | What kind of plan? Scoped implementation, phased feature, migration/rollout, architecture decision, or spec artifact. Ask only the questions that matter. |
| 19 | |
| 20 | ### 2. Ground in evidence |
| 21 | Read relevant files and search the codebase before proposing changes. Be explicit about what's confirmed from code vs. inferred vs. unknown. |
| 22 | |
| 23 | ### 3. Choose the right altitude |
| 24 | - **Small change** → concise plan with touchpoints and validation |
| 25 | - **Multi-file feature** → phased plan with dependencies |
| 26 | - **Architecture choice** → options, tradeoffs, recommendation |
| 27 | - **Migration / rollout** → sequencing, rollback, validation checkpoints |
| 28 | |
| 29 | Don't force every request into the same template. |
| 30 | |
| 31 | ### 4. Write the plan (right-size the artifact) |
| 32 | Match artifact to plan size — don't force `plans/*.md` on small scoped changes. |
| 33 | |
| 34 | - **Inline prose** — small single-file tweaks, quick scoped edits, brief decisions. Present the plan in the response and stop. |
| 35 | - **`plans/YYYYMMDDHHMMSS-{plan-name}.md`** — multi-file features, migrations, architecture decisions, anything that benefits from review or history. |
| 36 | - **`plans/YYYYMMDDHHMMSS-{plan-name}/plan.md` + phase files** — large multi-phase work where each phase warrants independent reading/editing. |
| 37 | |
| 38 | Use `date +%Y%m%d%H%M%S` for timestamps. |
| 39 | |
| 40 | A strong plan covers: problem summary, recommended approach, phases/workstreams, affected files/modules/systems, dependencies and sequencing, validation strategy, risks and mitigations, assumptions and open questions, non-goals when useful. |
| 41 | |
| 42 | ### 5. Present and stop |
| 43 | Summarize the recommendation, call out risks/assumptions/unknowns, clarify what needs user confirmation, then wait. Do not drift into coding. |
| 44 | |
| 45 | ## Boundaries |
| 46 | |
| 47 | - **Prose, not code** — describe *what* changes in prose (name the file, the concept); never include executable syntax (function bodies, JSX, SQL, migration scripts, shell commands). "Illustrative" snippets are still implementation code. |
| 48 | - **Right-sized artifacts** — don't force heavyweight structure onto small work; don't leave plans so abstract they name no touchpoints or validation path. |
| 49 | - **Every plan needs a validation path** — how will you know the implementation succeeded? |
| 50 | - Name concrete files, interfaces, and systems where possible. Surface tradeoffs instead of hiding them. |
| 51 | |
| 52 | ## Request |
| 53 | |
| 54 | <request>$ARGUMENTS</request> |