@brainervirus/workit-cursor 0.11.0 → 1.0.0

package/skills/workit-review/SKILL.md

@@ -0,0 +1,60 @@
+---
+name: workit-review
+description: Use when policy requires fresh-context review of a candidate or when an independent correctness and regression check is requested
+---
+
+# Review a candidate
+
+Review the real candidate in a stable context. A review is evidence about the
+current candidate, not an author's success summary.
+
+## Before method work
+
+If there is no active or paused task, run shared `task.start` then `policy.assess`
+before relying on selected policy rules or other product mutations. Assessment
+selects requirements; do not wait for a rule that can only exist after assess.
+
+## Method
+
+1. Pin or identify the candidate revision before reading conclusions. Inspect the
+   task objective, scope, constraints, accepted decisions, changed files, and
+   actual checks through shared `task`, `evidence`, and `policy` operations.
+2. Examine intent, correctness, regression risk, security or data consequences,
+   and project standards. Use the actual diff and check output; do not infer
+   evidence from a claim.
+3. Record each concern as a `finding` claim with its affected scope and candidate.
+   Investigate it: reproduce or trace the consequence, then fix in scope, dismiss
+   with evidence, defer with a reason, or ask the user about a real tradeoff.
+4. Reconcile conclusions when the candidate changes. Run one substantive review
+   and targeted rechecks; do not cycle reviewers indefinitely.
+
+If the required independent context is unavailable, record the review method as
+`unavailable` and preserve the gap. Same-session self-review is not independent
+review and must not be relabeled as verified.
+
+Use shared `evidence` and `finding` operations. Do not create a parallel review
+lifecycle, universal review panel, or direct metadata files.
+
+## Two axes, pinned
+
+Pin the fixed point first (`git diff <base>...HEAD` plus log); review that
+candidate only. Judge on two axes, never merged or reranked:
+
+- **Standards:** repo standards plus a smell baseline (mysterious name, long
+  method, duplicated logic, refused bequest, and kin); repo rules override
+  the baseline; judgement calls only, never tooling-enforced nits.
+- **Spec:** does the diff implement the originating spec/requirement
+  faithfully — missing, creep, or wrong, quoting the spec line.
+
+Every finding needs proof: the changed hunk, a failing/passing test ref, or
+a before/after. Causal disposition decides the outcome: introduced or
+worsened behavior gets fixed; pre-existing issues become follow-ups;
+inconclusive claims escalate, never silently pass.
+
+## Common mistakes
+
+| Mistake | Correction |
+| --- | --- |
+| Reviewing the summary instead of the candidate | Start from the stable candidate and real refs. |
+| Treating every comment as a defect | Investigate the claim and consequence first. |
+| Calling self-review independent | Preserve an unavailable capability gap. |

package/skills/workit-steer/SKILL.md

@@ -0,0 +1,32 @@
+---
+name: workit-steer
+description: Use when new instructions, interruptions, or forgotten items arrive mid-task
+---
+
+# Steer without losing the thread
+
+New context mid-session is normal; losing the thread is not. Park,
+classify, handle, re-anchor — every time.
+
+## Before method work
+
+If there is no active or paused task, run shared `task.start` then `policy.assess`
+before relying on selected policy rules or other product mutations. Assessment
+selects requirements; do not wait for a rule that can only exist after assess.
+
+## Method
+
+1. Park current state to task progress verbatim: summary, nextAction,
+   blockers. Never trust memory across an interruption.
+2. Classify the steering:
+   - same-task: fold into scope (reassess if facts changed), continue.
+   - new-task: `task.start` + `policy.assess`; the parked task waits.
+   - quick-question: answer from the parked state, then resume.
+3. Handle it with the same rigor as the parked work (no drive-by edits).
+4. Re-anchor: one-line resume brief (where we were, what changed, what
+   is next) before touching the parked work again.
+
+## Completion
+
+Both the steering and the parked work have an owner, a next action, and
+no silent drops. Interrupted work resumes from the brief, not from recall.

package/assets/templates/execution-contract.md

@@ -1,73 +0,0 @@
-Load `using-superpowers`, `subagent-driven-development`, `test-driven-development`, and `verification-before-completion` through OpenCode's `skill` tool. Implement the existing plan; do not re-plan.
-
-**Spec:** <SPEC_PATH>
-**Plan:** <PLAN_PATH>
-**Branch:** <BRANCH>
-**SDD:** `<SDD_DIR>`
-
-## Handoff destination
-
-This session is a handoff destination for a continued plan. The originating session already recorded the post-plan menu choice; present exactly these four choices plus model deferral and never re-offer the originating handoff option:
-
-- Subagent-driven
-- Inline
-- Review spec first
-- Review plan first
-- Change model first
-
-<workflow-handoff-destination>true</workflow-handoff-destination>
-
-## Hard gates
-
-- The parent is coordinator-only: it does not edit product code or perform delegated exploration.
-- Never use a worktree. Branch changes are in-place through `workit_branch_setup` on `feature/*` or `bugfix/*`; never commit on protected branches.
-- Working state, briefs, ledgers, and review diffs live only under gitignored `<SDD_DIR>` in `docs/<slug>/sdd/` and use `workit_sdd_*` tools.
-- Use native `todowrite` for visible task state as well as the gitignored ledger.
-- Use native `question` for branch/stash choices and guarded external mutations; call mutation tools only after approval with `confirmed: true` (grounded in the recorded NativeChoiceEvidence).
-- Flow-tool confirmations are never agent-typed booleans and never caller-supplied evidence objects: on OpenCode the plugin records the user's native-`question` answer as a host-observed one-use receipt (`attested: true`, `callID`, `selectedLabel`, `recordedAt`) consumed by `workit_spec_approve` / `workit_plan_approve` / `workit_plan_menu` — no evidence argument exists, and delegated worker status comes from host session parentage (`parentID`), never a caller `role` field. On Cursor, confirmations are policy-only (`attested: false`), Subagent-driven execution is supported through the one-time `coordinator_lease` and per-task `delegation_token` minted by `workit_delegate` (fail-closed validation; no `parentID` identity exists on Cursor), and Inline runs single-agent in the current session.
-- Delegated authority is direct-child-only: a worker is the session whose host `parentID` exactly equals the activating coordinator's recorded `coordinator_session_id`; missing, mismatched, or multi-owner lineage fails closed with `delegation_lineage_denied`, and nested `opencode` launches are denied during active delegated work. An authorized child receives only the compact worker contract (execute the supplied brief, follow TDD, land one contiguous non-empty commit range, report results) — never coordinator guidance, `wk-implement`, or ledger management; coordinator bookkeeping via `workit_sdd_*` stays with the coordinator session.
-- On Cursor, for every repository-scoped `workit_*` call, pass the active Cursor workspace as `workspace_root`; never rely on the MCP process default.
-- Use native `task` with only the built-in `explore` and `general` agents (OpenCode); on Cursor, Subagent-driven dispatches Cursor-native subagents with a task `delegation_token`, Inline runs single-agent in the current session.
-
-## Flow gates (HARD)
-
-- `wk-implement` refuses to run unless the plan is `approved` (flow.json) and the post-plan menu was presented.
-- `wk-handoff` refuses to run unless both spec and plan are `approved`.
-- Sequence is enforced by tools: `workit_spec_approve`, `workit_plan_approve`, `workit_plan_menu` — never skip a step (the spec/plan self-review runs automatically inside the transition; only the final approval asks for your confirmation).
-
-## Setup
-
-0. Call `workit_docs_validate` with the linked spec/plan paths. Hard-fail on any error before todos or branch setup.
-1. Call `workit_sdd_context` with `<PLAN_PATH>` and initialize `todowrite` from returned tasks.
-2. Call `workit_plan_tasks`; cache each top-level task's `section_text`.
-3. Mark IDs in `completed_task_ids` completed and never redispatch them.
-4. Call `workit_resolve_branch`, then show the current branch, target branch, and stash behavior before any in-place checkout/setup mutation.
-5. Always use `question`: for a clean tree ask whether to proceed or cancel; for a dirty tree add the stash choice and describe what will be stashed.
-6. Call `workit_branch_setup` with `confirmed: true` only after approval.
-
-## Remaining-task loop
-
-For each top-level task absent from `completed_task_ids`:
-
-1. Mark it `in_progress` with `todowrite`.
-2. Create a working-state brief with `workit_sdd_task_brief` and `confirmed: true`.
-3. Route by the approved execution mode: **Subagent-driven** delegates read-only discovery to an `explore` (OpenCode) agent, delegates implementation to a fresh `general` (OpenCode) agent (on Cursor, the coordinator mints a `delegation_token` with `workit_delegate` and the subagent prompt carries the raw token — the worker passes it as `delegation_token` on its mutation calls, and appending progress revokes it), and the coordinator never edits product code; **Inline** executes every task in the current agent with no dispatch and no token minting. Product changes follow TDD.
-4. Create a working-state diff with `workit_sdd_review_package` and `confirmed: true`.
-5. Delegate spec-compliance review and code-quality review to separate `general` agents.
-6. **Blocking** findings (Critical, Important, or spec-compliance) may trigger at most **two** fix+re-review rounds per task. **Advisory** findings (Minor, style, YAGNI, taste) never pause the loop — append them with `workit_sdd_append_advisory` (`--task <id> --text <text>`, `confirmed: true`) instead of an unrestricted file edit.
-7. Append the validated ledger entry with `workit_sdd_append_progress` and `confirmed: true`; mark the todo completed.
-
-## Final gate
-
-Run a separate full-branch code review, then `workit_verify`. Present the full `<SDD_DIR>/advisories.md` roll-up once, then use native `question` so the user can choose which advisory items to fix, discuss, or discard. Report exact check results and never infer success. Use `workit_git_context` for a commit preview and load `wk-commit` through `skill` for an approved commit. If working state contains a stash reference, preview reapplication through `question`, then call `workit_branch_setup` with `confirmed: true` after approval.
-
-**Mandatory:** end the run by calling `workit_plan_complete` (OpenCode/Cursor) or the CLI `workit flow complete` (CLI host) after the final task once the SDD ledger is complete (all task IDs appended) and `workit_verify` passes — a complete ledger and green verification are the tool's gates. Never finish the run while the plan is still `active`.
-
-## Task order
-
-<TASK_LIST>
-
-## Quality gate (HARD)
-
-- Specs/plans are written from `templates/spec-template.md` / `templates/plan-template.md`.
-- After `workit_docs_validate`, surface `quality` findings (spec scan). Hard findings (missing required section, missing CA-XX) block task start unless the user explicitly waives them. Warnings are advisory.

package/assets/templates/greeting.md

@@ -1 +0,0 @@
-{{greetingText}}

package/assets/templates/headers.md

@@ -1,3 +0,0 @@
-## Adjunto capturas
-
-## Archivos adjuntos

package/assets/templates/hygiene/.editorconfig

@@ -1,8 +0,0 @@
-root = true
-
-[*]
-charset = utf-8
-end_of_line = lf
-insert_final_newline = true
-indent_style = space
-indent_size = 2

package/assets/templates/hygiene/.gitattributes

@@ -1,3 +0,0 @@
-* text=auto
-*.md text
-*.bat text eol=crlf

package/assets/templates/hygiene/CHANGELOG.md

@@ -1,14 +0,0 @@
-# Changelog
-
-All notable changes to this project will be documented in this file.
-
-The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
-and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
-
-## [Unreleased]
-
-### Added
-
-### Changed
-
-### Fixed

package/assets/templates/hygiene/CONTRIBUTING.md

@@ -1,3 +0,0 @@
-# Contributing
-
-<!-- How to report issues, propose changes, and the review process. -->

package/assets/templates/hygiene/LICENSE

@@ -1,21 +0,0 @@
-MIT License
-
-Copyright (c) <YEAR> <HOLDER>
-
-Permission is hereby granted, free of charge, to any person obtaining a copy
-of this software and associated documentation files (the "Software"), to deal
-in the Software without restriction, including without limitation the rights
-to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
-copies of the Software, and to permit persons to whom the Software is
-furnished to do so, subject to the following conditions:
-
-The above copyright notice and this permission notice shall be included in all
-copies or substantial portions of the Software.
-
-THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
-IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
-FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
-AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
-LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
-OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
-SOFTWARE.

package/assets/templates/hygiene/README.md

@@ -1,3 +0,0 @@
-# <PROJECT>
-
-<!-- Describe the project: what it does, how to run it, how to contribute. -->

package/assets/templates/issue-update.md

@@ -1,6 +0,0 @@
-# Actualización
-{{greetingSection}}
-{{projectSection}}
-{{userNotesSection}}
-{{progressSection}}
-{{gitCommitsSection}}

package/assets/templates/plan-template.md

@@ -1,27 +0,0 @@
-# <Feature> Implementation Plan
-
-> **For agentic workers:** REQUIRED SUB-SKILL: Use superpowers:subagent-driven-development (recommended) or superpowers:executing-plans to implement this plan task-by-task. On Cursor the two paths are: **Subagent-driven** — Cursor-native subagents dispatched by the coordinator, each carrying a task-scoped `delegation_token` minted with `workit_delegate` from the one-time `coordinator_lease`; **Inline** — `executing-plans` in the current session, single-agent, no dispatch, no token minting. Steps use checkbox (`- [ ]`) syntax for tracking.
-
-**Spec:** `docs/<slug>/spec.md`
-**Branch:** `feature/<slug>`
-
-**Goal:** <one sentence>
-
-## Global Constraints
-
-- Each task lands exactly one contiguous non-empty commit range (`base..head`): fix rounds append commits to that range and never rewrite/amend an active review range; each progress line records the task's real base..head shas.
-- The final task ends execution with `workit_plan_complete` (or the CLI `workit flow complete`) once the SDD ledger is complete and repository verification passes — a run never finishes while the plan is still `active`.
-- <project-wide requirements, one line each>
-
----
-
-### Task N: <Component>
-
-- [ ] **Step 1: <action>**
-
-<!-- per-task criteria: how this task is verified -->
-**Criteria:** <verifiable check>
-
-| Status | Task |
-| --- | --- |
-| pending | N: <Component> |

package/assets/templates/spec-template.md

@@ -1,51 +0,0 @@
-# Spec: <feature>
-
-**Branch:** `feature/<slug>`
-
-## Context
-
-<!-- Why does this exist? What problem does it solve? 1-3 sentences. -->
-
-## Goals
-
-- <!-- measurable, one per bullet -->
-
-## Non-goals
-
-- <!-- explicitly out of scope -->
-
-## Architecture
-
-<!-- REQUIRED if this spec has flows or architecture: render a mermaid diagram (workit_present_flow). -->
-```mermaid
-flowchart TD
-  A[Start] --> B[Step]
-```
-
-<!-- REQUIRED if this spec touches UI: render an ASCII wireframe (workit_present_ascii). -->
-```text
-┌──────────────┐
-│ Header       │
-└──────────────┘
-```
-
-## Data flow / contracts
-
-<!-- REQUIRED when there is a glossary, scope comparison, or contracts: use markdown tables. -->
-| Term | Meaning |
-| --- | --- |
-| <term> | <meaning> |
-
-## Acceptance criteria
-
-<!-- REQUIRED: enumerable, each verifiable. Numbered CA-01, CA-02, ... -->
-- CA-01 …
-- CA-02 …
-
-## Decisions
-
-- D-01 …
-
-## Future work
-
-- …

package/assets/templates/superpowers-doc-contract.md

@@ -1,75 +0,0 @@
-# Superpowers document contract
-
-Use OpenCode's native `question` for every bounded user choice. Give concise choices and allow a custom answer; if the tool is unavailable, ask one concise plain-text question. Use `skill` to load workflow and Superpowers skills, `todowrite` for task state, and `task` for delegated work (OpenCode) or the host-native equivalent — on Cursor, Subagent-driven dispatches Cursor-native subagents with a task `delegation_token`, Inline runs single-agent in the current session.
-
-## Tracked document layout
-
-| Document | Path |
-| --- | --- |
-| Spec | `docs/<slug>/spec.md` |
-| Plan | `docs/<slug>/plan.md` |
-| SDD state | `docs/<slug>/sdd/` |
-
-An optional human mirror may live at `docs/specs/<same-basename>.md`; when present, keep its branch identical.
-
-Specs require:
-
-```markdown
-**Branch:** `feature/<slug>`
-```
-
-Plans require:
-
-```markdown
-**Spec:** `docs/<slug>/spec.md`
-**Branch:** `feature/<slug>`
-```
-
-`bugfix/<slug>` is also valid. Never use `main`, `develop`, `master`, or `prod`. Use plain backtick paths. Top-level headings are exactly `### Task N: Title`; steps use `- [ ] **Step N:** ...`; task headings never appear inside fences.
-
-Before writing **Branch:** into a new spec or plan, call `workit_docs_branch` and write the returned `branch` verbatim. When `action` is `keep`, use the current feature/bugfix branch. When `action` is `create_from_develop` or `create_from_base`, create the branch only through `workit_branch_setup`; it uses the configured workspace/global target branch.
-
-## Execution and handoff
-
-- Implementation uses `wk-implement` and subagent-driven development, with native `todowrite` and `task`.
-- Commits use `wk-commit` after its native `question` confirmation.
-- Continuation uses `wk-handoff`, whose `workit_handoff_session` creates and seeds the OpenCode session automatically.
-- Never use worktrees. Resolve the declared branch with `workit_resolve_branch`, preview dirty-tree stash choices with `question`, and apply an approved in-place checkout through `workit_branch_setup` with `confirmed: true` (grounded in the recorded NativeChoiceEvidence).
-- Flow-tool confirmations are never agent-typed booleans and never caller-supplied evidence objects: on OpenCode the plugin records the user's native-`question` answer as a host-observed one-use receipt (`attested: true`, `callID`, `selectedLabel`, `recordedAt`) consumed by `workit_spec_approve` / `workit_plan_approve` / `workit_plan_menu` — no evidence argument exists, and delegated worker status comes from host session parentage (`parentID`), never a caller `role` field. On Cursor, confirmations are policy-only (`attested: false`), Subagent-driven execution is supported through the one-time `coordinator_lease` and per-task `delegation_token` minted by `workit_delegate` (fail-closed validation; no `parentID` identity exists on Cursor), and Inline runs single-agent in the current session.
-- Keep all SDD state under the gitignored `docs/<slug>/sdd/`; use `workit_sdd_context` and the registered `workit_sdd_*` tools.
-- After implementation, use `question` before an approved stash reapply through `workit_branch_setup` with `confirmed: true`.
-
-## YouTrack content
-
-Chat follows the user's language. YouTrack task comments are Spanish (`es-CL`) and use `workit_youtrack_draft` followed by reviewed `workit_youtrack_post` with `confirmed: true`. Preserve the user's paragraph voice; do not inject commits, file paths, or robotic bullet reports.
-
-## Final self-check
-
-Before handoff, call `workit_docs_validate` on the linked spec/plan pair. Hard-fail on any error; never offer execution when validation fails.
-
-Before handoff, verify the saved spec path, plan path, declared branch, top-level task numbering, and workflow-managed SDD directory through the registered read-only workflow tools. Report structured failures; never infer success.
-
-## Post-plan execution choice
-
-After saving a plan, call `workit_docs_validate` on the spec/plan pair. On failure, stop and fix docs — do not offer execution.
-
-On success, use native `question` / Cursor `AskQuestion` with exactly these options (no stay, no A/B/C prose duplicate):
-
-1. Subagent-driven → load `wk-implement`
-2. Inline → execute in this session
-3. Handoff → load `wk-handoff` (new session only)
-4. Review spec first
-5. Review plan first
-6. Change model first
-
-`Change model first` is display-only deferral: it ends the turn without calling `workit_plan_menu` and re-presents the menu on the next turn. Every other choice must call `workit_plan_menu` immediately after the answer and before any skill, branch question, mutation, or handoff.
-
-Never emit Superpowers text beginning “Two execution options”.
-
-A handoff destination session (the seeded contract carries `<workflow-handoff-destination>true</workflow-handoff-destination>`) presents exactly five choices — Subagent-driven, Inline, Review spec first, Review plan first, Change model first — and never re-offers the originating handoff option.
-
-- Specs/plans must follow `templates/spec-template.md` / `templates/plan-template.md` (mandated diagrams, tables, CA-XX).
-
-## Doc delivery
-
-When delivering a spec or plan, use a clickable markdown link (\`[spec.md](docs/<slug>/spec.md)\`) and a 3-5 bullet summary of the content (Context, Goals, key decisions, status). Never reference docs with backtick-only paths.

package/rules/ask-question-only.mdc

@@ -1,49 +0,0 @@
----
-description: ALL user choices must use Cursor AskQuestion — never A/B/C in chat
-alwaysApply: true
----
-
-# AskQuestion only (HARD-GATE)
-
-**Overrides** Superpowers brainstorming “ask in chat / A/B/C text options” and any skill that suggests conversational multiple choice.
-
-## Required path for every decision
-
-When the user must choose (clarifying questions, product identity, layout, yes/no, confirm, branch, stash, YouTrack, approaches, or how to execute a completed plan):
-
-1. Call Cursor's native **`AskQuestion`** with a workflow-specific title, prompt, and options.
-2. Wait for the tool result — do **not** also paste the options in the chat message.
-3. If native `AskQuestion` is unavailable in the current mode, ask one concise plain-text question.
-
-## Forbidden (always)
-
-- Letter options in markdown: `A)`, `B)`, `C)`, `Reply with A, B, or C`
-- Bullet / numbered choice lists meant as the answer UI
-- “First clarifying question” blocks that list options in prose
-- Asking the user to type yes/no when `yes_no` + AskQuestion exists
-
-## Mapping
-
-| Need | Kind |
-|------|------|
-| 2+ custom choices (brainstorming, identity, layout, …) | `pick_option` + title, prompt, `id:label` args |
-| Yes / no | `yes_no` |
-| Branch / stash / YouTrack / VCS | named kinds from the tool |
-| Plan completed; choose execution method | `plan_execution` + saved plan path |
-
-One AskQuestion per turn when possible. Open-ended free text (no options) may stay in chat — **as soon as there are options, AskQuestion is mandatory**.
-
-## Superpowers writing-plans override
-
-After writing a plan, **do not emit** Superpowers' inline text beginning “Two execution options”.
-
-After `workit_docs_validate` succeeds on the saved spec/plan pair, call Cursor `AskQuestion` with exactly:
-
-- Subagent-driven → dispatches Cursor-native subagents: `workit_plan_menu` returns a one-time `coordinator_lease`, `workit_delegate` mints a task-scoped `delegation_token` per task, and the worker passes it as `delegation_token` on mutation calls
-- Inline → execute in this session
-- Handoff → `/wk-handoff` (new session only)
-- Review spec first
-- Review plan first
-- Cursor's built-in **Other** response, always last
-
-No `--stay` option in this menu.

package/rules/cursor-todowrite.mdc

@@ -1,14 +0,0 @@
----
-description: SDD ledger AND Cursor TodoWrite — both required for plan execution
-alwaysApply: true
----
-
-# TodoWrite UI (HARD-GATE)
-
-When executing a plan / SDD handoff:
-
-1. MCP `workit_sdd_context` returns `todos` — call Cursor **TodoWrite** with that array (`merge: false`) before Task 1.
-2. SDD under `docs/<slug>/sdd/` is persistence only — **not** a substitute for the Cursor task list UI.
-3. Per task: `in_progress` before dispatch; `completed` after `workit_sdd_append_progress`.
-
-Never skip TodoWrite because the SDD ledger exists.

package/rules/no-worktrees.mdc

@@ -1,26 +0,0 @@
----
-description: NEVER use git worktrees — in-place feature/* or bugfix/* checkout only
-alwaysApply: true
----
-
-# No worktrees (HARD-GATE)
-
-**Overrides** Superpowers `using-git-worktrees` skill and any execution contract that mentions worktrees.
-
-## Policy
-
-- **NEVER** invoke `using-git-worktrees`
-- **NEVER** run `git worktree add` or any worktree command
-- Branch work is **in-place only**: `feature/*` or `bugfix/*`
-- Never commit on `main`, `develop`, `master`, or `prod`
-
-## Required flow (before Task 1)
-
-1. MCP `workit_resolve_branch` with spec + plan paths
-2. If `needs_checkout` and `dirty` → native **AskQuestion** asks whether to stash before checkout
-3. MCP `workit_branch_setup` with `target_branch` and `stash: yes|no`
-4. After all tasks, if manifest has `stash_ref` → AskQuestion → `workit_branch_setup` action=`reapply_stash`
-
-## Tools only
-
-Do not infer branch or stash state from manual `git status`. Use MCP tools.

package/rules/sdd-docs-path.mdc

@@ -1,22 +0,0 @@
----
-description: SDD artifacts live under docs/<slug>/sdd — never .superpowers/sdd
-alwaysApply: true
----
-
-# SDD path (HARD-GATE)
-
-**Overrides** Superpowers `subagent-driven-development` ledger check at `.superpowers/sdd/progress.md`.
-
-## Required
-
-1. MCP `workit_sdd_context` with `plan_path` — **always**, including fresh projects
-2. Use returned `sdd_dir` (`docs/<slug>/sdd/`) for all SDD artifacts
-3. `workit_sdd_task_brief`, `workit_sdd_review_package`, `workit_sdd_append_progress` — never Superpowers `sdd-workspace` script
-
-## Forbidden
-
-- `.superpowers/sdd/` for progress, briefs, or review diffs
-- `cat .superpowers/sdd/progress.md` — missing file there does NOT mean "start fresh"
-- Superpowers `task-brief` / `review-package` scripts under `.agents/skills/`
-
-Fresh project with no ledger → `workit_sdd_context` creates nothing; `progress.md` appears only on the first confirmed `workit_sdd_append_progress`.

package/skills/wk-changelog/SKILL.md

@@ -1,75 +0,0 @@
----
-name: wk-changelog
-description: Update CHANGELOG.md via workit_changelog_context + workit_changelog_apply. Use for /wk-changelog or "update the changelog".
----
-
-# Changelog — Keep a Changelog Update
-
-Update `CHANGELOG.md` for the current repository.
-
-## Step 1 — Gather facts (required)
-
-Call MCP tool `workit_changelog_context` with arguments from the user's message (range, version, paths, etc.).
-
-**Repository calls:** For every repository-scoped `workit_*` call, pass the active Cursor workspace as `workspace_root`; never rely on the MCP process default.
-
-Use the tool return value as ground truth. Do not read git, run npm, or infer repo state yourself.
-If the tool errors, report the error and stop.
-
-If `unreleased.needs_normalize` is true, call `workit_changelog_apply` with `normalize_only: true` before merging new bullets.
-
-## Step 2 — Decide entries (do not edit the file yet)
-
-Follow Keep a Changelog 1.1.0:
-
-- Target `## [Unreleased]` only (unless the user asks to cut a version).
-- Categories: `Added`, `Changed`, `Deprecated`, `Removed`, `Fixed`, `Security`.
-- Human-readable bullets, not raw commit messages.
-- Group by user-visible behavior, not by file.
-- Skip internal refactors / CI / tests / formatting unless they affect users.
-- If there are no meaningful user-visible changes, stop — do not invent entries.
-
-## Step 3 — Apply via tool (required)
-
-Call MCP `workit_changelog_apply` with the new bullets. Examples:
-
-```json
-{
-  "entries": {
-    "Added": ["**Foo API** — public endpoint for bar"],
-    "Fixed": ["**Login redirect** — stop loop on expired session"]
-  }
-}
-```
-
-or
-
-```json
-{
-  "entries": [
-    { "category": "Added", "text": "**Foo API** — public endpoint for bar" }
-  ]
-}
-```
-
-### HARD-GATE
-
-- **NEVER** hand-edit `CHANGELOG.md` to append a new `### Added` / `### Changed` / … block under `[Unreleased]`.
-- The apply tool **merges** into the existing category heading (or creates one category once).
-- Duplicate identical bullets are skipped.
-- Do not update package versions or tags.
-- Do not commit.
-
-## Output
-
-After the tool returns ok, summarize:
-
-```md
-Changelog update:
-
-- <entry or reason no update was needed>
-
-Files changed:
-
-- CHANGELOG.md
-```