shape
/shape
Use when a feature has settled intent and codebase facts but multiple viable implementation approaches remain.
- Author
- ebullient ↗
- Repository
- qrspi-x ↗
- Licence
- Apache-2.0
- Last repository change
- 23.10.2026
Assessment and security
- Files in this package
- 1
- Scripts
- 0
- HTML comments
- 0
- URLs in SKILL.md
- 0
No signals detected. Repository changed on 23.10.2026, without references to retired models or APIs.
Apache-2.0, otwarta: można używać i zmieniać, z zachowaniem licencji i informacji o zmianach.
What it does and when to use it
Use when a feature has settled intent and codebase facts but multiple viable implementation approaches remain.
Original author description
Use when a feature has settled intent and codebase facts but multiple viable implementation approaches remain.
How it works
- Core Philosophy
- The helper
- Task
- After the Agent Returns
SKILL.md preview
--- name: shape description: 'Use when a feature has settled intent and codebase facts but multiple viable implementation approaches remain.' when_to_use: 'Use optionally after Query ↔ Research and before Spec when the solution direction is not obvious. Skip it and use `qrspi-x:spec` when one reasonable approach is clear; use `qrspi-x:query` or `qrspi-x:research` when intent or codebase facts are still unclear.' disable-model-invocation: false compatibility: Node 22+ --- # QRSPI Shape ## Core Philosophy - Resolve meaningful implementation alternatives before freezing the behavioral contract This skill is part of the QRSPI workflow and is normally invoked by `qrspi-x:workflow`. It may also be invoked directly. Runtime contract: read `../workflow/references/runtime.md` before dispatch. It defines the fresh-context, capability, fallback, and on-disk evidence requirements; this skill adds only Shape-specific inputs and output. ## The helper Helper installation, recovery, and artifact-only fallback are defined by `qrspi-x:workflow`. Shape has no `qrspi-x` state of its own; when the helper is available it's used only for backup naming on a rerun, per the Task steps below. `status`'s optional `current.step` convenience reads `approach.md`'s `## Decision` section, but never gates on it. If it exits 127, continue this interactive step without it. Never edit any helper state manually. ## Task Explore and compare candidate implementation approaches for a settled feature. Shape is an optional, human-gated definition step: it decides how the feature should be approached, but it does not write product code, define the behavioral spec, or create an implementation plan. The shaper uses all available pre-definition context: - `./qrspi/<feature>/request.md` — the settled intent - `./qrspi/<feature>/background.md` — optional human context and prior art - `./qrspi/<feature>/queries.md` — the questions that framed research - `./qrspi/<feature>/research.md` — codebase facts and constraints 1. Verify that `request.md`, `queries.md`, and `research.md` exist. If any is missing, stop and direct the caller to Init, Query, or Research. `background.md` is optional. 2. If `research.md` has a non-empty `## New Questions`, stop and return to Query before shaping; do not design around unresolved facts. 3. If `approach.md` exists, retain its complete contents and move it to `./qrspi/<feature>/backups/approach-<n>.md`, where `n` is one greater than the highest `n` already present for the `approach` stem, starting at 1 (if the helper is available, `qrspi-x next-file approach --feature <feature> --project <path>` returns this path directly — same result, no need to list `backups/` and compute `n` by hand). Never rename, rotate, or overwrite an existing backup — writing one is always a pure addition. Create `backups/` only when there is something to put in it. 4. Before spawning, prefer the declared agent when the runtime supports named agents: - If `qrspi-x:shaper` is registered, spawn it directly so the runtime can apply its declared settings. - Otherwise, read `../../agents/shaper.md`, resolved relative to this `SKILL.md`, and spawn a general-purpose subagent with its full contents as the role instructions. 5. Pass the feature name so the agent can locate the four artifacts, noting that `background.md` may be absent. The agent may read the codebase to validate candidate approaches, but it must write only `./qrspi/<feature>/approach.md`. 6. If `approach.md` already exists, preserve its prior decision and rationale while refining the alternatives; never silently replace a recorded decision. The approach artifact should make the choice reviewable: ```markdown # QRSPI Approach: <feature> ## Context ... ## Constraints and Facts ... ## Candidate Approaches ### Option A: ... ... ## Recommendation ... ## Decision None. ## Open Questions ... ``` ## After the Agent Returns 1. Read the agent's report and inspect `approach.md` for candidate approaches, tradeoffs, and a recommendation. 2. If the comparison exposes missing facts, stop and return to Query/Research rather than guessing. 3. Stop for human review. The human may approve the recommendation, select another option, or edit `approach.md`; replace `None.` under `## Decision` with the selected approach and rationale before proceeding to Spec. The shaping pass is complete when `approach.md` is written, but the human decision is a separate gate. 4. Leave `None.` in place under `## Decision` until the human decides — any other text, including a comment, reads as a decision. `status`'s optional `current.step` convenience reads this same section to tell `shape` from `spec`, but never gates on it; the Shape gate itself stays this skill's own read of `approach.md`. Do not write `spec.md` or plan files. If the approach is obvious, skip this step entirely and use `qrspi-x:spec`.
Versions
| Repository | Score | Licence |
|---|---|---|
| This page · ebullient/qrspi-x | 100 | Apache-2.0 |
| biliboss/skills | 100 | MIT |
| gmdkaio/kstack | 100 | MIT |
| YnotnaUK/ai-skills | 100 | MPL-2.0 |
| jamesm-dev/payday-dev | 100 | MIT |
| nyelonong/galdr | 100 | MIT |
| thecraighewitt/skills | 100 | MIT |
| tiago-peixoto/claude-shapeup | 95 | MIT |
| educlopez/ui-craft | 92 | MIT |
Database snapshot: 13.11.2026 · package version 6ec106caa9a9 (SHA-256 digest of the entire package)
Downloads in the last 30 days: 0 · ranking
Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.
Similar skills
- 100research
Investigate a codebase, compare tools or approaches, verify current external or library facts, and research failures before making a substantive technical recommendation. Use for sufficiency and gap reviews; ordinary edits with settled requirements do not need a research stage. - 90brainstorming
Draws out what the user wants and why, compares two or three approaches grounded in the code, grills open decisions in rounds, and gets a design doc approved before any code is written. Use before building or changing anything whose intent is not settled: a new feature,… - 100write-plan
Write an implementation plan. Use when the behavior is settled and how to build or fix it is open — a pinned cause, settled spec, hunt finding, or review comment. Not for deciding the behavior (write-specs) or multi-session work (frontier). - 90compare-options
Compare multiple approaches, libraries, or patterns. Searches codebase for existing usage, checks memory for past experiences, creates comparison matrix. Use when deciding between approaches before implementation. - 90design
Brainstorm design options for a feature and document the result. Use when the user asks to design a new feature or a major change to an existing one, compare implementation approaches, or create/update a design doc in docs/design/ or experiments/<username>/<project>/docs. - 100capture-decisions
Capture settled durable decisions into existing owning repo memory, or ask where to record them before creating a destination. Use after discussion, research, specs, reviews, or implementation when the user asks to record decisions, durable memory, glossary terms, ADR-worthy…