docs: align skill workflow documentation

skills/create-plan/cursor/SKILL.md

@@ -43,7 +43,7 @@ If any dependency is missing, stop and return:
 
 ## Required Skill Invocation Rules
 
-- Invoke relevant skills through workspace discovery (`.cursor/skills/`).
+- Invoke relevant skills through Cursor-native discovery (`.cursor/skills/`, `~/.cursor/skills/`, or installed Cursor plugin cache entries).
 - Announce skill usage explicitly:
   - `I've read the [Skill Name] skill and I'm using it to [purpose].`
 - For skills with checklists, track checklist items explicitly in conversation.

skills/do-task/cursor/SKILL.md

@@ -43,7 +43,7 @@ If any required dependency is missing, stop immediately and return:
 
 ## Required Skill Invocation Rules
 
-- Invoke relevant skills through workspace discovery (`.cursor/skills/` repo-local or `~/.cursor/skills/` global).
+- Invoke relevant skills through Cursor-native discovery (`.cursor/skills/`, `~/.cursor/skills/`, or installed Cursor plugin cache entries).
 - Announce skill usage explicitly:
   - `I've read the [Skill Name] skill and I'm using it to [purpose].`
 - For skills with checklists, track checklist items explicitly in conversation.
@@ -91,7 +91,7 @@ If any required dependency is missing, stop immediately and return:
    - Append `/ai_plan/` to `.gitignore`.
    - Commit that infra change immediately with message `chore(gitignore): ignore ai_plan local planning artifacts`.
    - This infra commit is EXPLICITLY separate from the task commit in Phase 9. It may occur even when the final task ends up `aborted` or `failed`.
-3. Verify required sub-skills are discoverable under `.cursor/skills/superpowers/skills/` or `~/.cursor/skills/superpowers/skills/`. Announce each sub-skill before invocation using:
+3. Verify required sub-skills are discoverable through Cursor-native skill discovery: the Cursor Superpowers plugin cache, `.cursor/skills/superpowers/skills/`, or `~/.cursor/skills/superpowers/skills/`. Announce each sub-skill before invocation using:
    `I've read the [Skill Name] skill and I'm using it to [purpose].`
 
 ### Phase 2: Parse Prompt and Question
@@ -107,7 +107,7 @@ If any required dependency is missing, stop immediately and return:
      - All identifiers in the prompt are resolvable against the codebase.
    - Otherwise, ask 1-3 clarifying questions, ONE AT A TIME, multiple-choice preferred.
    - Empty prompt → ask exactly once: "what task?".
-5. Invoke `superpowers:brainstorming` via workspace discovery (`.cursor/skills/superpowers/skills/brainstorming/SKILL.md`) for any **behavior-changing** task — feature creation, bug fix with multiple plausible approaches, refactor, design decision. Present 2-3 approaches and recommend one before finalizing the plan. The ONLY skip conditions are the same ones that allow TDD auto-skip: `pure-documentation` and `pure-comment-whitespace-rename`. When skipping, record the skip reason in the Interpretation section of `task-plan.md`.
+5. Invoke `superpowers:brainstorming` through Cursor-native skill discovery for any **behavior-changing** task — feature creation, bug fix with multiple plausible approaches, refactor, design decision. Present 2-3 approaches and recommend one before finalizing the plan. The ONLY skip conditions are the same ones that allow TDD auto-skip: `pure-documentation` and `pure-comment-whitespace-rename`. When skipping, record the skip reason in the Interpretation section of `task-plan.md`.
 
 ### Phase 3: Configure Reviewer
 
@@ -163,7 +163,7 @@ Steps:
    - Leave `Runtime State`, `Review History`, `Final Status` empty (skill updates these).
 6. Set `Status: draft`.
 
-**Worktree branch:** If the prompt opts in to a worktree (see Trigger Phrase Detection), invoke `superpowers:using-git-worktrees` via workspace discovery before proceeding. Otherwise continue on the current branch.
+**Worktree branch:** If the prompt opts in to a worktree (see Trigger Phrase Detection), invoke `superpowers:using-git-worktrees` via Cursor-native discovery before proceeding. Otherwise continue on the current branch.
 
 ### Phase 5: Plan Review Loop
 
@@ -224,7 +224,7 @@ Native orchestration — do not invoke `superpowers:executing-plans`.
 
 1. Set `Status: implementation-in-progress`.
 2. For every behavior-changing file edit:
-   - Invoke `superpowers:test-driven-development` via workspace discovery.
+   - Invoke `superpowers:test-driven-development` via Cursor-native discovery.
    - Write the failing test first. Run it. Confirm it fails.
    - Implement the minimal code to make it pass. Run the test. Confirm green.
    - Do NOT commit yet — a single task commit happens in Phase 9.
@@ -236,7 +236,7 @@ Native orchestration — do not invoke `superpowers:executing-plans`.
 
 ### Phase 7: Verification Gate
 
-Invoke `superpowers:verification-before-completion` via workspace discovery.
+Invoke `superpowers:verification-before-completion` via Cursor-native discovery.
 
 Run the commands listed in the `Verification` section of `task-plan.md`:
 - Lint (changed files first).
@@ -326,7 +326,7 @@ On MAX_ROUNDS:
 
 ### Phase 9: Commit + Push Ask
 
-Invoke `superpowers:finishing-a-development-branch` via workspace discovery.
+Invoke `superpowers:finishing-a-development-branch` via Cursor-native discovery.
 
 1. Stage all changed files explicitly (avoid `git add -A`).
 2. Single commit with message derived from the task goal:
@@ -803,11 +803,11 @@ Review History is append-only.
 
 ## Variant Hardening Notes — Cursor
 
-- Must use workspace discovery from `.cursor/skills/` (repo-local) or `~/.cursor/skills/` (global).
+- Must use Cursor-native discovery from `.cursor/skills/`, `~/.cursor/skills/`, or installed Cursor plugin cache entries.
 - Reviewer invocation MUST use `--mode=ask --trust --output-format json`. Never `--mode=agent`, never `--force`, never write-capable modes for reviewer calls.
 - `jq` is a hard prerequisite because the cursor reviewer branch always returns JSON that must be parsed for `session_id` and `result`.
 - Helper paths are `.cursor/skills/reviewer-runtime/...` preferred, with `~/.cursor/skills/reviewer-runtime/...` as the fallback path.
-- Sub-skills are invoked through workspace discovery with explicit announcement — Cursor does not expose a `Skill` tool.
+- Sub-skills are invoked through Cursor-native discovery with explicit announcement — Cursor does not expose a `Skill` tool.
 - `cursor-agent --version` must succeed as part of the prerequisite check. The binary is `cursor-agent`; the alias `cursor agent` is equivalent.
 - No plan-mode guard (Cursor Agent CLI has no plan-mode concept).
 

skills/do-task/opencode/SKILL.md

@@ -9,7 +9,7 @@ Execute an ad-hoc user prompt end-to-end: parse → clarify → plan (with revie
 
 This is a single-artifact sibling of `create-plan` + `implement-plan`. Unlike `implement-plan`, `do-task` operates on one persistent `task-plan.md` (not a full milestone plan) and defaults to the **current branch** (not a worktree).
 
-**Core principle:** OpenCode loads skills through its native skill tool. Local skills live under `~/.config/opencode/skills/`, and OpenCode can also expose shared agent skills from `~/.agents/skills/`. Sub-skill invocations use OpenCode's native mechanism — not Claude's `Skill` tool, not Cursor's workspace discovery.
+**Core principle:** OpenCode loads skills through its native skill tool. Local skills live under `~/.config/opencode/skills/`, and OpenCode can also expose shared agent skills from `~/.agents/skills/`. Sub-skill invocations use OpenCode's native mechanism — not Claude's `Skill` tool, not Cursor's discovery mechanism.
 
 ## Prerequisite Check (MANDATORY)
 
@@ -45,7 +45,7 @@ If any required dependency is missing, stop immediately and return:
 - Announce skill usage explicitly:
   - `I've read the [Skill Name] skill and I'm using it to [purpose].`
 - For skills with checklists, track checklist items explicitly in conversation.
-- Do NOT use Claude's `Skill` tool syntax or Cursor's workspace discovery. OpenCode's skill system may expose shared files from `~/.agents/skills/`, but invocation still goes through OpenCode's native skill mechanism.
+- Do NOT use Claude's `Skill` tool syntax or Cursor's discovery mechanism. OpenCode's skill system may expose shared files from `~/.agents/skills/`, but invocation still goes through OpenCode's native skill mechanism.
 
 ## Trigger Phrase Detection
 

skills/implement-plan/cursor/SKILL.md

@@ -57,7 +57,7 @@ If no plan folder exists:
 
 ## Required Skill Invocation Rules
 
-- Invoke relevant skills through workspace discovery (`.cursor/skills/`).
+- Invoke relevant skills through Cursor-native discovery (`.cursor/skills/`, `~/.cursor/skills/`, or installed Cursor plugin cache entries).
 - Announce skill usage explicitly:
   - `I've read the [Skill Name] skill and I'm using it to [purpose].`
 - For skills with checklists, track checklist items explicitly in conversation.