qrspi-plan
Skill3 starsChanged 4 months agoMITView on GitHub
QRSPI Plan phase -- converts an approved spec.md into a mechanically executable, vertically sliced plan.md with exact file paths and per-phase verification. Use for "/qrspi-plan <feature>", "qrspi plan X", "turn the spec into an implementation plan". Not for planning without a spec; this phase consumes spec.md and REFUSES horizontal-layer plans.
In this repository86 other files it gives its agents.
- AGENTS.mdAGENTS.md
- project-templates/AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- project-templates/CLAUDE.mdCLAUDE.md
- alembic-migration-managerSkill
- architecture-journalSkill
- architecture-reviewSkill
- automated-code-reviewSkill
- axum-scaffolderSkill
- capture-consolidateSkill
- cargo-package-scaffoldSkill
- code-review-coachSkill
- confluence-guide-writerSkill
- dependency-mapperSkill
- doc-syncSkill
- dotnet-architecture-checklistSkill
- dotnet-controller-api-scaffolderSkill
- dotnet-security-reviewSkill
- dotnet-vertical-sliceSkill
- ef-migration-managerSkill
- email-captureSkill
- environment-healthSkill
- evaluate-testsSkill
- fastapi-scaffolderSkill
- fitness-functionsSkill
- jira-comment-writerSkill
- legacy-migration-analyzerSkill
- minimal-api-scaffolderSkill
- nuget-package-scaffoldSkill
- oss-vettingSkill
- pattern-tradeoff-analyzerSkill
- php-api-scaffolderSkill
- php-architecture-checklistSkill
- php-feature-sliceSkill
- php-migration-managerSkill
- php-package-scaffoldSkill
- php-security-reviewSkill
- pr-feedback-writerSkill
- pypi-package-scaffoldSkill
- python-architecture-checklistSkill
- python-feature-sliceSkill
- python-modernization-analyzerSkill
- python-security-reviewSkill
- qraspi-architectureSkill
- qraspi-graduateSkill
- qraspi-implementSkill
- qraspi-planSkill
- qraspi-questionsSkill
- qraspi-researchSkill
- qraspi-skeletonSkill
- qrspi-implementSkill
- qrspi-questionsSkill
- qrspi-researchSkill
- qrspi-specSkill
- react-app-scaffolderSkill
- react-architecture-checklistSkill
- react-component-scaffolderSkill
- react-feature-sliceSkill
- react-modernization-analyzerSkill
- react-security-reviewSkill
- The first 60 shown.
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
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 31
EVERY PHASE VERIFIABLE -- each phase carries an automated command (build / test / lint) and a rollback
- Distinctive: line 32
NO SOURCE CODE -- change descriptions and illustrative snippets only, never full implementations
≈ “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
QRSPI Plan
"A plan precise enough to execute mechanically is the goal. If the implementer has to invent anything, the plan failed." -- Adapted from Dwight D. Eisenhower
Core Philosophy
The Plan phase turns the approved spec into an implementation contract the Implement phase can execute WITHOUT judgment, invention, or discovery. The spec already settled the design and the vertical slices; Plan makes each slice mechanical -- exact file paths, exact changes, a test-first step, and an automated verification command. Its one hard gate is vertical, not horizontal: a plan organized by technical layer (all models, then all services, then all UI) is rejected before it is written. Each phase must deliver an end-to-end testable increment, inherited from the spec's slices.
Non-Negotiable Constraints:
- SPEC-GATED -- never plan without an approved
spec.mdon disk; never plan from memory - VERTICAL OR REFUSE -- if phases organize by horizontal layer, STOP and re-slice before writing
- MECHANICAL EXECUTION -- every change names an exact file path; "the service layer" is not a path
- TESTS FIRST -- any phase adding behavior opens with a failing-test step (RED) before the code step
- EVERY PHASE VERIFIABLE -- each phase carries an automated command (build / test / lint) and a rollback
- NO SOURCE CODE -- change descriptions and illustrative snippets only, never full implementations
- CONTEXT BUDGET: keep utilization under 40%. At 60%, write
plan.mdwith progress and tell the user to start a fresh session.
Workflow
PRE-FLIGHT
[ ] Locate the feature folder thoughts/shared/qrspi/YYYY-MM-DD-{slug}/
[ ] Read spec.md. If absent -> STOP; route the user to /qrspi-spec
[ ] Confirm spec.md status: approved AND design_approved: true.
If only ready-for-review -> STOP; ask the human to approve the spec design first
[ ] Carry the spec's Structure Outline slices forward as the phase skeleton
RE-SLICE GATE (the vertical-not-horizontal refusal)
Inspect the intended phases. If any phase completes a whole layer (all models, then all
services, then all UI), STOP and re-slice so each phase is an end-to-end increment.
Do NOT proceed to WRITE until every phase is a vertical slice.
WRITE (only after the gate passes)
For each slice/phase: exact file paths · REMOVE/ADD change descriptions · a RED test step
before the code step · an automated verification command · a rollback line.
Add "What we're NOT doing" (scope boundaries) and a rollback plan.
Write plan.md, status: ready-for-review.
REPORT
Artifact path · phase list (one line each) · "Review/approve, then start a NEW session and run /qrspi-implement"Exit criteria: plan.md holds vertically sliced phases, each with exact paths, a test-first
step, an automated verification command, and a rollback; a "What we're NOT doing" list is present;
status: ready-for-review; user told to review/approve before /qrspi-implement.
State Block
<qrspi-plan-state>
phase: PRE-FLIGHT | RE-SLICE-GATE | WRITE | REPORT | COMPLETE
feature_folder: thoughts/shared/qrspi/YYYY-MM-DD-{slug}/
spec_present: true | false # MUST be true to proceed
spec_approved: true | false # MUST be true to proceed
phases_planned: [count]
vertical_check: pass | re-slice-needed # MUST be pass before WRITE
tests_first_phases: [count]
rollback_documented: true | false
context_budget: under-40 | approaching-60 | checkpoint-now
status: draft | ready-for-review | approved
</qrspi-plan-state>Output Template
See references/plan-template.md for the full plan.md structure, frontmatter, per-phase shape
(exact paths, RED/GREEN test steps, verification commands, rollback), and the horizontal-vs-vertical
worked example.
Integration with Other Skills
| Skill | Relationship |
qrspi-spec | Prior phase. Its approved spec.md (vertical slices + signatures) is the contract this phase makes mechanical. |
qrspi-implement | Next phase. Consumes the approved plan.md; refuses to execute without status: approved. |
tdd-loop | The inner loop the Implement phase runs per phase; Plan's test-first steps map onto RED-GREEN-REFACTOR. |
dotnet-vertical-slice / python-feature-slice / rust-feature-slice | Stack scaffolders for the vertical slices each phase implements. |
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.

