macca-method 1.0.0 → 2.0.0

package/.agents/legacy-payloads.json

@@ -0,0 +1,22 @@
+{
+  "1.1.0": {
+    "_shared": "fa7d1f79c95bfe33cffd2ea0215e7bc3c01aaf1d5a72d8462a626f5c7ff8396e",
+    "add-feature": "dd31b4cbf66e09b313fdfa2d47d05a9625d5b68f265ddf0e2d9264f9d6309f4d",
+    "brainstorm-api": "07349d8b7ff9b9dadbb89a5dcfd1bfdba1e40e14b913bf94ef4f33ecc24987fa",
+    "brainstorm-architecture": "97193d976c5cecdf5cc12dde4d0b691ecaf4fe1fd23bf8252dbc7e8bdd161359",
+    "brainstorm-prd": "ac5c7cd95f33083f030cf2a0b2a1db8a9833fb5da3db11b72e729c96d36d58b0",
+    "brainstorm-rules": "4a1ddd8c7e36eb6abd481af276ac6d6eac183431f2eceaf902e7cb33b0eba744",
+    "brainstorm-schema": "06f6251ac07583d4118c8bdb88a39c812ba390a81a487b04ddf830f4bd0dd83a",
+    "brainstorm-styleguide": "2f0fc76c074097ae94e54e17deda18209c8d6c39053c35150ed953b5cf0af9b7",
+    "brainstorm-task": "2c0cb3a51b42ab9a2784f668473535d07bd46eb099768af9de000a42912596c6",
+    "bug-fix": "fae7c87df1c50abd0361806e6e4e472827939de87407eaff1485665fa35b1aa6",
+    "code-review": "df924d268b3068cfab2f8f3b1046bdbc9341d69cf62b2c7c2f8951ef6a3b0f39",
+    "developer": "9e446d83747c86609d7dff34e258d3e83d8b6267fcbb91723f861fbe8311e084",
+    "help": "7cb8ebe46069266e56fee5cacbc5c7bc28f68459e8752a1a1ed5a8734c436f06",
+    "quick-dev": "4912623f5f7c2697dadaebbd04e950b5afb08301246279fa1eb948aeedcb9371",
+    "base64:cmFwYXQ=": "630692d55e8bf515fd0fe263404cce034d66d9b7037dec26df8a3017626a6473",
+    "spec-audit": "397a44c512c02ff186ad1855c9040c4b735b8d5f578b69bc78d4831b62e76728",
+    "spec-compliance": "0ca7aff9bbdc7109d529dfe4ed6024bacc56df6ee028b92f1d52be6ce931ae04",
+    "spec-init": "30466716e0f88e4b8511c7797a95306f1575345c691daf9fa7331726cfcb47b4"
+  }
+}

package/{skills-lock.json → .agents/macca-lock.json}

@@ -1,5 +1,5 @@
 {
-  "version": "1.0.0",
+  "version": "2.0.0",
   "skills": [
     "_shared",
     "add-feature",
@@ -14,7 +14,9 @@
     "code-review",
     "developer",
     "help",
-    "rapat",
+    "meet",
+    "quick-dev",
+    "release-readiness",
     "spec-audit",
     "spec-compliance",
     "spec-init"

package/.agents/macca-managed-skills.txt

@@ -11,7 +11,9 @@ bug-fix
 code-review
 developer
 help
-rapat
+meet
+quick-dev
+release-readiness
 spec-audit
 spec-compliance
-spec-init
+spec-init

package/.agents/skills/_shared/references/additional-skills.md

@@ -0,0 +1,30 @@
+# Additional Skills Compatibility
+
+Readers of `.agents/developer-config.json` must tolerate all supported `additionalSkills` forms.
+
+Canonical writer form:
+
+```json
+{
+  "name": "frontend-react-best-practices",
+  "purpose": "Use when working on React UI code.",
+  "paths": {
+    "copilot": ".github/skills/frontend-react-best-practices/SKILL.md",
+    "opencode": ".opencode/skills/frontend-react-best-practices/SKILL.md",
+    "codex": ".agents/skills/frontend-react-best-practices/SKILL.md"
+  }
+}
+```
+
+Readers must also accept:
+
+- `path`
+- `githubPath`, `opencodePath`, `claudePath`, `cursorPath`, `windsurfPath`, `geminiPath`, `kiloPath`, `kimiPath`, `codexPath`
+
+Path fallback order:
+
+1. `paths[currentHost]`
+2. `path`
+3. matching legacy host-specific field
+
+Do not remove legacy fields during unrelated config updates.

package/.agents/skills/_shared/references/brainstorm-session.md

@@ -1,10 +1,20 @@
 # Brainstorm Session Policy
 
+## Table of Contents
+
+1. Purpose
+2. Shared Preferences
+3. Discovery Depth
+4. Interview Pace
+5. Recommendations Toggle
+6. Session Setup Rules
+7. Recommended Prompt Patterns
+
 ## Purpose
 
 This file defines shared session behavior for all `brainstorm-*` skills.
 
-The caller skill must read `runtime-config.md` before this file.
+The caller skill must read `language-config.md` and `config-mutation.md` before using persisted preferences.
 
 ## Shared Preferences
 
@@ -17,13 +27,33 @@ Currently supported fields:
 {
   "brainstormPreferences": {
     "discussionMode": "one-by-one",
-    "recommendations": true
+    "recommendations": true,
+    "discoveryDepth": "standard"
   }
 }
 ```
 
 Accepted `discussionMode` values: `"one-by-one"`, `"three-at-a-time"`, `"all-at-once"`.
 
+Accepted `discoveryDepth` values: `"quick"`, `"standard"`, `"critical"`.
+
+## Discovery Depth
+
+Depth controls detail, not whether mandatory safety topics are skipped:
+
+- `quick` — prototype or small internal experiment; concise answers and fewer follow-ups. Use only when the user explicitly identifies the project as disposable/prototype work.
+- `standard` — default production depth.
+- `critical` — deeper evidence, failure, security, recovery, and operational detail for payments, sensitive/regulated data, multi-tenancy, public uploads/webhooks, privileged administration, high availability, or material financial/legal impact.
+
+Determine depth from existing evidence before asking:
+
+1. Start from a saved valid value if present.
+2. If current evidence shows critical-risk signals, escalate the active depth to `critical` even when a saved value is `quick` or `standard`.
+3. Select `quick` only from explicit user intent for disposable prototype/internal work.
+4. Otherwise use `standard`.
+
+Announce the selected depth with the saved pacing. The user may override it, but do not add a separate setup question merely to choose depth.
+
 ## Interview Pace
 
 Brainstorm skills may batch questions when the user explicitly wants faster progress. This is separate from approval gates in execution skills.
@@ -50,20 +80,21 @@ Use `brainstormPreferences.recommendations` as a persistent preference:
 
 When a brainstorming skill starts:
 
-1. Read `languagePreferences` via `runtime-config.md`.
+1. Read `languagePreferences` via `language-config.md`.
 2. Read `brainstormPreferences` if present.
 3. **Announce the session** before asking anything:
-   - If preferences are already saved — show a short confirmation and allow changes:
-     ```
-     This session has [N] topics.
-     Saved preferences: [pacing] | recommendations: [on/off]
-     Continue with these settings? Or type the changes you want.
-     ```
+   - If preferences are already saved, announce and proceed in the same response. The user may override them at any time:
+      ```
+      This session has [N] topics.
+      Saved preferences: [pacing] | recommendations: [on/off] | depth: [quick/standard/critical]
+      Using these settings. Type different settings at any time.
+      ```
    - If no preferences are saved — ask both before starting:
      ```
      This session has [N] topics. Two things before we start:
      1. Pace: (A) one by one  (B) three at a time  (C) all at once
-     2. Answer recommendations: should the AI suggest answers for each question? (Y/N)
+      2. Answer recommendations: should the AI suggest answers for each question? (Y/N)
+      Depth defaults to standard and increases automatically for sensitive systems. Type quick/standard/critical only to override it.
      ```
 4. Save the chosen preferences. Preserve all unrelated config fields when writing the update.
 
@@ -72,7 +103,7 @@ When a brainstorming skill starts:
 For pace:
 
 ```text
-This session has [N] topics. Do you want to discuss them one by one or three at a time?
+This session has [N] topics. Choose one by one, three at a time, or all at once.
 ```
 
 For recommendations:

package/.agents/skills/_shared/references/config-mutation.md

@@ -0,0 +1,25 @@
+# Config Mutation Contract
+
+`.agents/developer-config.json` is a stable public runtime contract.
+
+Before writing it:
+
+1. Read the existing object if present.
+2. Merge only the intended fields.
+3. Preserve every unknown and unrelated field.
+4. Never replace the whole object unless the file does not exist.
+5. Prefer additive schema changes. Do not silently rename or repurpose keys.
+
+Stable fields include:
+
+- `name`
+- `project`
+- `languagePreferences`
+- `developerPreferences.workMode`
+- `developerPreferences.scope`
+- `brainstormPreferences`
+- `additionalSkills`
+- `availableMCPs`
+- `codeReviewPreferences.fixMode`
+
+If migration becomes unavoidable, readers must remain backward compatible until the installer can migrate old values automatically.

package/.agents/skills/_shared/references/finding-format.md

@@ -0,0 +1,25 @@
+# Shared Finding Format
+
+Use exactly these four points for every actionable finding. Do not add a fifth point and do not show code/diffs inside the four points.
+
+```markdown
+#### [Severity] [ID] [Short Title]
+
+**Where?**
+[Page or file name only]
+
+**What happens if it is not fixed?**
+[Real user/application consequence in clear language]
+
+**What happens if it is fixed?**
+[Practical benefit]
+
+**Recommended fix**
+[Required logic/flow change, not syntax]
+```
+
+Rules:
+
+- Keep severity proportional to evidence and impact.
+- Do not create empty findings for passing checks.
+- Put exact technical targets and validation in the separate fix manifest.

package/.agents/skills/_shared/references/fix-mode.md

@@ -0,0 +1,39 @@
+# Fix Mode and Approval Resume Contract
+
+`codeReviewPreferences.fixMode` is binding for review and remediation workflows.
+
+## Values
+
+- Missing or `report-first`: analyze and report all findings, then stop at the gate before editing.
+- `fix-then-report`: apply actionable BLOCKER/MAJOR fixes, validate them, then report. Document-audit skills edit only when corrections were explicitly requested.
+
+Announce `[Fix mode: report-first]` or `[Fix mode: fix-then-report]` at the start of a new review. Do not repeat the announcement when resuming an active gate.
+
+## Required Report-First Gate
+
+After the full report and fix manifest, print this block verbatim and end the response:
+
+```text
+---
+[GATE — Fix mode: report-first]
+All findings have been reported. No files were changed.
+Reply "yes" / "fix" / "continue" to apply all fixes,
+or name which findings you want to fix.
+---
+```
+
+The fix manifest records each actionable finding ID, target, bounded change, and validation.
+
+## Approval Resume
+
+This takes precedence over normal routing, startup, identity, onboarding, preflight, and work-mode prompts.
+
+1. Trim and case-fold the next response.
+2. Exact `yes`, `fix`, or `continue` approves every actionable finding in the immediately preceding report, including actionable MINOR findings.
+3. Named IDs approve only those findings.
+4. Resume directly at edits. Do not repeat analysis, the report, startup, preflight, or the same gate.
+5. Ask again only if the worktree changed materially, a target disappeared, findings conflict, or the implementation becomes destructive or exceeds disclosed scope.
+6. Run the narrowest relevant validation and one bounded verdict pass: `resolved`, `partial`, or `unresolved`.
+7. If validation still fails after one repair pass, stop and report evidence.
+
+`quick-dev` and other routers must not intercept a reply to an active gate.

package/.agents/skills/_shared/references/human-loop.md

@@ -52,4 +52,6 @@ Keep one decision topic per pause.
 
 ## Resume Rule
 
-After the user answers, continue from the exact paused step. Do not restart the workflow or ask for the same confirmation again unless the situation changes.
+After the user answers, continue from the exact paused step. Do not restart the workflow or ask for the same confirmation again unless the situation materially changes.
+
+An answer to an active gate takes precedence over normal skill startup and routing. For a report-first gate, exact `yes`, `fix`, or `continue` and finding-ID subsets follow the Approval Resume Protocol in `fix-mode.md`. Do not send them through identity setup, preflight, or a second confirmation.

package/.agents/skills/_shared/references/implementation-principles.md

@@ -0,0 +1,19 @@
+# Shared Implementation Principles
+
+Before writing code, stop at the first sufficient option:
+
+1. Do not build what is not needed.
+2. Search for and reuse existing project behavior.
+3. Prefer the standard library.
+4. Prefer native platform/framework capability.
+5. Reuse an installed dependency.
+6. Use the smallest clear implementation.
+7. Only then write new reusable code.
+
+Additional rules:
+
+- Comments explain why, not what.
+- Fix root causes, not symptoms.
+- Prefer deletion and direct code over speculative abstraction.
+- Business behavior or scope changes require user/spec approval; low-risk technical implementation choices do not.
+- A new dependency requires evidence that earlier steps are insufficient and follows the project's dependency policy.

package/.agents/skills/_shared/references/invocation-policy.md

@@ -0,0 +1,39 @@
+# Skill Invocation Policy
+
+This catalog documents intent; it does not replace host routing or safety gates.
+
+Portable Agent Skills rely on `name` and `description`. OpenCode ignores Copilot-specific invocation fields. Therefore canonical skills keep standard frontmatter and express routing through descriptions and workflow boundaries.
+
+| Skill | Policy | Rationale |
+|---|---|---|
+| `add-feature` | explicit-intent | Broad source-of-truth mutation |
+| `brainstorm-prd` | explicit-intent | Long interview and PRD creation |
+| `brainstorm-architecture` | explicit-intent | Strategic technical decisions |
+| `brainstorm-schema` | explicit-intent | Consequential data contract |
+| `brainstorm-api` | explicit-intent | External contract creation |
+| `brainstorm-styleguide` | explicit-intent | UI contract interview |
+| `brainstorm-rules` | explicit-intent | Repository-wide rules |
+| `brainstorm-task` | both | Direct use and `add-feature` orchestration |
+| `spec-init` | explicit-intent | Whole-codebase scan and multi-document generation |
+| `developer` | explicit-implementation-intent | Broad implementation mutation |
+| `quick-dev` | model-auto | Bounded router for small implementation tasks |
+| `bug-fix` | both | Automatic bug/error routing and direct use; the first code change still requires explicit implementation approval from the diagnosis gate |
+| `spec-compliance` | orchestrated | Called by execution/remediation workflows; ad hoc requests remain valid |
+| `code-review` | both | Direct review and post-phase orchestration |
+| `spec-audit` | both | Direct audit and final workflow check |
+| `release-readiness` | both | Direct pre-release request and project-complete handoff |
+| `help` | both | Cheap read-only routing |
+| `meet` | both | Non-mutating meeting intent |
+
+Definitions:
+
+- `explicit-intent`: activate only when the request clearly asks for that workflow; the user need not type the skill name.
+- `explicit-implementation-intent`: implementation wording such as "implement Phase 2" is sufficient.
+- `model-auto`: choose automatically from context.
+- `orchestrated`: primarily loaded by another skill.
+- `both`: direct and automatic/orchestrated use are valid.
+
+Host adapters may improve discoverability but must not carry essential behavior:
+
+- Copilot VS Code: `disable-model-invocation: true` for explicit-only; `user-invocable: false` for model/orchestrated-only.
+- OpenCode: use descriptions, skill permissions, and optional custom commands; it has no equivalent portable frontmatter fields.

package/.agents/skills/_shared/references/language-config.md

@@ -0,0 +1,15 @@
+# Language Config Contract
+
+Read `.agents/developer-config.json` if it exists.
+
+Use:
+
+- `languagePreferences.communication.normalized` for chat, reports, prompts, and confirmations.
+- `languagePreferences.documents.normalized` for generated project documents.
+
+Accepted reader values:
+
+- Indonesian: `indonesian`, `id`
+- English: `english`, `en`
+
+If missing, use Indonesian. Never translate file names, traceability IDs, config keys, or code literals.

package/.agents/skills/_shared/references/output-ownership.md

@@ -14,17 +14,19 @@ This file defines which skill primarily owns each persistent output, so skills d
 | `project-context/api.md` | `brainstorm-api` | `add-feature`, `spec-init` |
 | `project-context/StyleGuide.md` | `brainstorm-styleguide` | `add-feature`, `spec-init` |
 | `project-context/rules.md` | `brainstorm-rules` | `add-feature`, `spec-init` |
-| `project-context/Task.md` | `brainstorm-task` | `developer` only updates progress; `add-feature` must hand off to `brainstorm-task` |
+| `project-context/Task.md` | `brainstorm-task` | `developer` updates progress and approved phase deltas; `quick-dev` adds one approved lightweight entry; `add-feature` hands off to `brainstorm-task` |
 | `project-context/bug-log.md` | `bug-fix` | none |
 | `project-context/plans/*.md` | `developer` | `add-feature`, `code-review` limited updates per workflow |
 
 ## Ownership Rules
 
 - `help` explains and routes. It does not create project spec files.
-- `rapat` facilitates decisions and maps them to target artifacts. It does not replace the primary owner of spec files.
+- `meet` facilitates one structured persona round and maps decisions to target artifacts. It does not replace the primary owner of spec files.
 - `spec-audit` and `spec-compliance` report findings. They do not rewrite spec artifacts unless the user explicitly asks for follow-up fix work.
 - `code-review` reports code issues. It does not own spec documents.
 - `spec-init` is the bootstrap exception for existing codebases without specs.
+- `bug-fix` may add one narrowly scoped regression guard to a spec or rule only after user confirmation; larger document changes return to the primary owner.
+- `release-readiness` is report-only and owns no persistent project artifact unless the user explicitly requests a saved copy of its report.
 
 ## Drift Rule
 

package/.agents/skills/_shared/references/runtime-config.md

@@ -1,171 +1,10 @@
-# Runtime Config Contract
+# Runtime Config Reference Index
 
-## Table of Contents
+This compatibility index replaces the former monolithic runtime contract. Skills should load only the references needed for their current workflow:
 
-1. Purpose
-2. Required Read Order
-3. Language Preferences
-4. Stable Shared Fields
-5. Fix Mode Contract
-6. `additionalSkills` Compatibility
-7. Mutation Rules
+- Language selection: `language-config.md`
+- Safe config writes and stable fields: `config-mutation.md`
+- Review gates and approval resume: `fix-mode.md`
+- Additional skill path compatibility: `additional-skills.md`
 
-## Purpose
-
-This file is the source of truth for how MACCA skills read and update `.agents/developer-config.json`.
-
-Treat `developer-config.json` as a stable public contract. Do not make breaking schema changes that force existing users to edit the file manually after an upgrade.
-
-## Required Read Order
-
-Before any user-facing output or config mutation:
-
-1. Read `.agents/developer-config.json` if it exists.
-2. Preserve unknown fields when writing updates.
-3. Merge changes into the existing object. Never replace the whole file unless the file does not exist yet.
-
-## Language Preferences
-
-Use this field if present:
-
-```json
-{
-  "languagePreferences": {
-    "communication": {
-      "raw": "Indonesian",
-      "normalized": "indonesian"
-    },
-    "documents": {
-      "raw": "Indonesian",
-      "normalized": "indonesian"
-    }
-  }
-}
-```
-
-### Accepted Compatibility Values
-
-Readers must accept long and short normalized forms:
-
-- Indonesian: `indonesian`, `id`
-- English: `english`, `en`
-
-Writers should preserve known values when possible. If a skill must write a new normalized value, use the installer's canonical form unless the installer is updated to a new canonical form first.
-
-### Output Rules
-
-- Use `languagePreferences.communication.normalized` for chat output, reports, prompts, and confirmations.
-- Use `languagePreferences.documents.normalized` for generated project documents.
-- Never translate file names, traceability IDs, config keys, or code literals.
-
-## Stable Shared Fields
-
-These fields are part of the stable contract and must remain backward compatible:
-
-- `name`
-- `project`
-- `languagePreferences`
-- `developerPreferences.workMode`
-- `developerPreferences.scope`
-- `brainstormPreferences`
-- `additionalSkills`
-- `availableMCPs`
-- `codeReviewPreferences.fixMode`
-
-## Fix Mode Contract
-
-`codeReviewPreferences.fixMode` is a **binding setting for review/remediation skills**.
-`developer`, `bug-fix`, `spec-compliance`, `code-review`, and `spec-audit` MUST follow this field before they fix findings or change files in a review/remediation workflow.
-
-Brainstorming/document-generation skills such as `brainstorm-*`, `add-feature`, and `spec-init` are NOT required to use `fixMode`, because they do not run a report-findings-then-fix workflow.
-
-### Read and Announce at Startup
-
-Every covered review/remediation skill must read `fixMode` during Shared Runtime Setup — **before any analysis or action begins**:
-
-1. Read `codeReviewPreferences.fixMode` from `developer-config.json`.
-2. If it is missing, treat it as `"report-first"` — do not ask the user.
-3. Announce it at the top of the output: `[Fix mode: report-first]` or `[Fix mode: fix-then-report]`.
-
-### Values
-
-| Value | Behavior |
-|-------|----------|
-| `"report-first"` | **Default.** Run all analysis. Present the full findings report. Show the gate prompt below. **End the response.** Wait for user confirmation in the next message before touching any files. |
-| `"fix-then-report"` | Automatically apply BLOCKER/MAJOR fixes. Present the full report at the end. |
-
-### Required Gate Prompt (`report-first` only)
-
-After presenting all findings, the skill must display this block **verbatim**, then **end the response immediately**:
-
-```
----
-[GATE — Fix mode: report-first]
-All findings have been reported. No files were changed.
-Reply "yes" / "fix" / "continue" to apply all fixes,
-or name which findings you want to fix.
----
-```
-
-**Critical rule:** DO NOT apply any fixes, edit any files, run any sub-skill, or add follow-up text in the same response. The response ends at the gate prompt. Act only after the user's next message confirms.
-
-### Default
-
-If `codeReviewPreferences.fixMode` is missing from `developer-config.json`, always treat it as `"report-first"`. Do not ask the user — just use the default and announce it.
-
-## `additionalSkills` Compatibility
-
-> **Internal contract only.** These forms exist for backward compatibility across different AI hosts (Copilot, OpenCode, Codex, etc.), each of which stores skill files in different paths. Users do not choose the form — skills always write the canonical extensible form when saving. Readers must tolerate all three forms.
-
-Readers must support all of these forms:
-
-### Legacy Single-Path Form
-
-```json
-{
-  "name": "frontend-react-best-practices",
-  "path": ".agents/skills/frontend-react-best-practices/SKILL.md",
-  "purpose": "Use when working on React UI code."
-}
-```
-
-### Legacy Host-Specific Form
-
-```json
-{
-  "name": "frontend-react-best-practices",
-  "opencodePath": ".opencode/skill/frontend-react-best-practices/SKILL.md",
-  "githubPath": ".github/skills/frontend-react-best-practices/SKILL.md",
-  "purpose": "Use when working on React UI code."
-}
-```
-
-### Canonical Extensible Form
-
-```json
-{
-  "name": "frontend-react-best-practices",
-  "purpose": "Use when working on React UI code.",
-  "paths": {
-    "copilot": ".github/skills/frontend-react-best-practices/SKILL.md",
-    "opencode": ".opencode/skill/frontend-react-best-practices/SKILL.md",
-    "codex": ".agents/skills/frontend-react-best-practices/SKILL.md"
-  }
-}
-```
-
-### Read Fallback Order
-
-When selecting a skill path for the current host, use this order:
-
-1. `paths[currentHost]`
-2. `path`
-3. legacy host-specific fields such as `githubPath`, `opencodePath`, `claudePath`, `cursorPath`, `windsurfPath`, `geminiPath`, `kiloPath`, `kimiPath`, `codexPath`
-
-Do not remove legacy fields during unrelated updates.
-
-## Mutation Rules
-
-- Additive changes only by default.
-- Do not silently rename or repurpose existing keys.
-- If a future migration is unavoidable, skills must remain tolerant readers until upgrade tooling can migrate old configs automatically.
+Do not load every contract by default. Existing external references to `runtime-config.md` remain valid through this index, but canonical MACCA skills should point to the narrower file directly.

package/.agents/skills/_shared/references/skill-catalog.md

@@ -0,0 +1,34 @@
+# MACCA Skill Contract Catalog
+
+Compact framework-audit index. This is not a replacement for each `SKILL.md`; it identifies which full files need inspection when a conflict is suspected.
+
+| Skill | Invocation | Reads | Writes / Output | Primary Handoff |
+|---|---|---|---|---|
+| `brainstorm-prd` | explicit-intent | scope/config, user discovery | `PRD.md` | `brainstorm-architecture` |
+| `brainstorm-architecture` | explicit-intent | PRD | `architecture.md` | schema/API/style/rules |
+| `brainstorm-schema` | explicit-intent | PRD, architecture | `schema.md` | `brainstorm-api` |
+| `brainstorm-api` | explicit-intent | PRD, architecture, schema when applicable | `api.md` | style/rules/task |
+| `brainstorm-styleguide` | explicit-intent | PRD, architecture | `StyleGuide.md` | rules/task |
+| `brainstorm-rules` | explicit-intent | architecture and applicable domain specs | `rules.md` | `brainstorm-task` |
+| `brainstorm-task` | both | applicable specs | `Task.md` | `developer` |
+| `add-feature` | explicit-intent | all existing specs | approved spec deltas; delegates Task | `developer` |
+| `spec-init` | explicit-intent | existing code/config/evidence | bootstrap specs, not Task | audit/task |
+| `developer` | explicit-implementation-intent | Task and relevant specs | code, Task progress, plans | compliance/review |
+| `quick-dev` | model-auto | relevant specs | bounded code and Task entry | compliance/review |
+| `bug-fix` | both | bug log, relevant code/specs | approved fix, prevention, confirmed bug log | compliance/review |
+| `spec-compliance` | orchestrated | code and applicable specs | report; approved fixes | code-review |
+| `code-review` | both | diff/code, rules, architecture, applicable contracts | report; approved fixes | phase completion |
+| `spec-audit` | both | spec pairs or framework catalog | report; explicitly requested corrections | owning skills |
+| `release-readiness` | both | existing quality/release evidence | report only | release owner |
+| `help` | both | project status/config/catalog | routing report | selected skill |
+| `meet` | both | agenda-relevant evidence | one-round discussion report | owning skills |
+
+Framework invariants:
+
+- Persistent artifact ownership comes from `output-ownership.md`.
+- Scope comes from `scope-rules.md`.
+- Review gates come from `fix-mode.md`.
+- Invocation intent comes from `invocation-policy.md`.
+- Brainstorm pacing/depth comes from `brainstorm-session.md`.
+- `release-readiness`, `help`, and `meet` are report-only.
+- Execution order is `developer/quick-dev/bug-fix` → `spec-compliance` → `code-review`.