Exécutez n'importe quel Skill dans Manus
en un clic

Exécutez n'importe quel Skill dans Manus en un clic

cursorqb-implementer

Étoiles23

Forks2

Mis à jour14 juin 2026 à 09:19

Gated, goal-backed Step 4 of the CursorQB workflow - implement one bounded, reversible, testable slice from an audited plan. Use after the Step 3 audit when the user approves implementation, or via /cursorqb-implement. Runs only when the audit is PASS or PASS_WITH_WARNINGS with no P0/P1 findings; otherwise it stops and recommends repair. Selects a single READY sub-plan, defines the validation command first, makes a minimal change, and verifies before claiming done.

Installation

Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.

Exécuter dans Manus

Source

alicankiraz1

alicankiraz1/CursorQB

Ouvrir le dépôt GitHub Voir les dépôts du créateur

Téléchargement

Exécuter dans Manus

Explorateur de fichiers

2 fichiers

SKILL.md

readonly

Plus depuis ce dépôt

même dépôt

cursorqb-auditor

alicankiraz1/CursorQB

Goal-backed Step 3 of the CursorQB planning workflow - a quality, coverage, consistency, and readiness audit of the Step 2 sub-plans. Use to verify that Planner-docs/Faz-*-Plans/*.md and Sub-Planing-Index.md are faithful to Main-Planing.md, complete, ordered, well-structured, and ready for implementation. Launched automatically as a Cursor goal via the define-goal skill and runs until every phase and sub-plan is inspected; produces only Planner-docs/Sub-Planing-Audit.md and returns PASS / PASS_WITH_WARNINGS / BLOCKED. Safe to run standalone for re-audits.

2026-06-1423

cursorqb-autopsy

alicankiraz1/CursorQB

Goal-backed Step 1.5 of the CursorQB planning workflow - an existing-project autopsy. Use after Step 1 (or directly) to inspect an existing/partially built repository and write Planner-docs/Autopsy.md, a 13-section technical feedback report that enriches Step 2 sub-plans. Runs automatically for existing/non-empty repositories and is skipped for empty or nearly empty ones. Launched automatically as a Cursor goal via the define-goal skill; only creates/updates Planner-docs/Autopsy.md.

2026-06-1423

cursorqb-planner

alicankiraz1/CursorQB

Guided five-step (1, 1.5, 2, 3, 4), goal-backed project planning orchestrator with a repo-aware intake. Use when the user runs /cursorqb-plan or asks CursorQB to plan a project end to end - master plan, existing-project autopsy, phase decomposition, coverage audit, then optional implementation. Runs a repo-aware Step 1 intake, produces Planner-docs/Main-Planing.md, runs a Step 1.5 autopsy (cursorqb-autopsy) for existing projects, then gates into phase decomposition (cursorqb-subplanner), a coverage/quality audit (cursorqb-auditor), and a gated implementation step (cursorqb-implementer) - launched automatically as Cursor goals via define-goal. Validates each step with the bundled validator.

2026-06-1423

cursorqb-subplanner

alicankiraz1/CursorQB

Goal-backed Step 2 of the CursorQB planning workflow. Use to decompose every main phase in Planner-docs/Main-Planing.md into detailed sub-plans under Planner-docs/Faz-<n>-Plans/ plus Sub-Planing-Index.md. Launched automatically as a Cursor goal via the define-goal skill and runs until ALL phases are covered - it does not stop after one phase. Invoke from cursorqb-planner Gate 1, or directly to (re)generate sub-plans. Only changes files under Planner-docs/.

2026-06-1423

name

cursorqb-implementer

description

CursorQB Implementer (Step 4, gated, goal-backed)

Implement one bounded, reversible slice from the audited CursorQB plan. This is the only CursorQB step that changes source code, and it runs strictly behind the audit gate. Read the full Step-4 spec in fourth-planner.md (next to this skill) and follow it end to end.

Execution model

In-session, goal-backed. Read the bundled fourth-planner.md spec and follow it. Register the slice goal via the Cursor define-goal skill (get_goal then create_goal); if the goal tool is unavailable, proceed under the in-context goal contract - behavior is identical.
Read the reliability notes. Resolve the plugin root by walking up to the folder containing .cursor-plugin/plugin.json, then read references/workflow-quality.md (especially Step 4 Token Discipline) before starting.
Scope is the user's workspace (cwd). All reads/edits target the user's active workspace, never this plugin directory.

Gate (mandatory, before any change)

Run the bundled validator in Step 4 mode (resolve <plugin-root> as above):

python3 <plugin-root>/scripts/validate_planner_docs.py --root . --mode step4

BLOCKED audit status -> do not implement; surface the blocker and minimal unblock action; stop.
Any P0 or P1 finding -> do not implement; recommend the targeted repairs first (cursorqb-auditor fix list + cursorqb-planner repair loop); stop.
PASS, or PASS_WITH_WARNINGS with only P2/P3 -> proceed and keep the warnings visible.
python3 unavailable -> read the audit's ## 1, ## 12, ## 13, ## 15 sections and apply the same gate manually.

Token discipline

Read Planner-docs/Sub-Planing-Audit.md and Planner-docs/Sub-Planing-Index.md first.
Select ONE READY or READY_WITH_WARNINGS sub-plan from the audit's ## 12. Step 4 Readiness Assessment table (highest-priority first, unless the user names another).
Load only that sub-plan and the repo files needed for the slice. Do not load all sub-plans.

Slice procedure

Read git status, README.md, AGENTS.md, Makefile, the audit, and the selected sub-plan.
Determine the validation/test command FIRST; prefer existing repo commands over invented ones; write a failing test first when feasible.
Make the minimal, reversible change for the slice.
Run the focused tests plus the relevant make smoke/check target.
Verify with fresh evidence before claiming done.
Report an exact blocker or success; separate code-delivery status from external config/credential blockers.

Leverage installed skills (optional, with fallback)

If installed/available, use these by scope; if not installed, do not stop - continue with the audit, the selected sub-plan, repo instructions, and existing validation commands using the same principles:

Superpowers executing-plans or subagent-driven-development for execution structure.
Superpowers test-driven-development for code changes.
Superpowers verification-before-completion before asserting the slice is done.
The security review (review-security / security-review) for security-, policy-, secret-, or command-execution-sensitive changes.

Stop rules

One sub-plan and one reversible slice per run.
Never write secrets, tokens, private keys, or local credentials.
Do not commit, push, open a PR, deploy, or mutate external systems unless the user explicitly asks in this run.
Do not claim success without running the validation command and confirming its output.

Output

Report: which sub-plan slice was implemented, the validation command and its result, files changed, remaining P2/P3 warnings to keep visible, and the recommended next slice or the exact blocker. After a successful slice, the user can run /cursorqb-implement again for the next READY sub-plan.