opencode-multiagent 0.2.1 → 0.4.0

package/skills/sadd-judge-with-debate/SKILL.md

@@ -20,7 +20,7 @@ This command implements iterative multi-judge debate:
 
 ```
 Phase 0: Setup
-         mkdir -p .specs/reports
+         mkdir -p .opencode/reports
                   │
 Phase 1: Independent Analysis
          ┌─ Judge 1 → {name}.1.md ─┐
@@ -46,12 +46,13 @@ Phase 2: Debate Round (iterative)   │
 Before starting evaluation, ensure the reports directory exists:
 
 ```bash
-mkdir -p .specs/reports
+mkdir -p .opencode/reports
 ```
 
-**Report naming convention:** `.specs/reports/{solution-name}-{YYYY-MM-DD}.[1|2|3].md`
+**Report naming convention:** `.opencode/reports/{solution-name}-{YYYY-MM-DD}.[1|2|3].md`
 
 Where:
+
 - `{solution-name}` - Derived from solution filename (e.g., `users-api` from `src/api/users.ts`)
 - `{YYYY-MM-DD}` - Current date
 - `[1|2|3]` - Judge number
@@ -64,7 +65,7 @@ Launch **3 independent judge agents in parallel** (recommended: Opus for rigor):
    - Path to solution(s) being evaluated
    - Evaluation criteria with weights
    - Clear rubric for scoring
-2. Each produces **independent assessment** saved to `.specs/reports/{solution-name}-{date}.[1|2|3].md`
+2. Each produces **independent assessment** saved to `.opencode/reports/{solution-name}-{date}.[1|2|3].md`
 3. Reports must include:
    - Per-criterion scores with evidence
    - Specific quotes/examples supporting ratings
@@ -91,12 +92,13 @@ You are Judge {N} evaluating a solution independently.
 </evaluation_criteria>
 
 <output_file>
-.specs/reports/{solution-name}-{date}.{N}.md
+.opencode/reports/{solution-name}-{date}.{N}.md
 </output_file>
 
-Read ${CLAUDE_PLUGIN_ROOT}/tasks/judge.md for evaluation methodology and execute using following criteria.
+Apply the evaluation methodology criteria below.
 
 Instructions:
+
 1. Read the solution thoroughly
 2. For each criterion:
    - Find specific evidence (quote exact text)
@@ -121,8 +123,8 @@ For each debate round (max 3 rounds):
 Launch **3 debate agents in parallel**:
 
 1. Each judge agent receives:
-   - Path to their own previous report (`.specs/reports/{solution-name}-{date}.[1|2|3].md`)
-   - Paths to other judges' reports (`.specs/reports/{solution-name}-{date}.[1|2|3].md`)
+   - Path to their own previous report (`.opencode/reports/{solution-name}-{date}.[1|2|3].md`)
+   - Paths to other judges' reports (`.opencode/reports/{solution-name}-{date}.[1|2|3].md`)
    - The original solution
 2. Each judge:
    - Identifies disagreements with other judges (>1 point score gap on any criterion)
@@ -141,11 +143,11 @@ Launch **3 debate agents in parallel**:
 You are Judge {N} in debate round {R}.
 
 <your_previous_report>
-{path to .specs/reports/{solution-name}-{date}.{N}.md}
+{path to .opencode/reports/{solution-name}-{date}.{N}.md}
 </your_previous_report>
 
 <other_judges_reports>
-Judge 1: .specs/reports/{solution-name}-{date}.1.md
+Judge 1: .opencode/reports/{solution-name}-{date}.1.md
 ...
 </other_judges_reports>
 
@@ -158,12 +160,13 @@ Judge 1: .specs/reports/{solution-name}-{date}.1.md
 </solution_path>
 
 <output_file>
-.specs/reports/{solution-name}-{date}.{N}.md (append to existing file)
+.opencode/reports/{solution-name}-{date}.{N}.md (append to existing file)
 </output_file>
 
-Read ${CLAUDE_PLUGIN_ROOT}/tasks/judge.md for evaluation methodology principles.
+Apply the evaluation methodology principles below.
 
 Instructions:
+
 1. Read your previous assessment from {your_previous_report}
 2. Read all other judges' reports
 3. Identify disagreements (where your scores differ by >1 point)
@@ -182,6 +185,7 @@ Instructions:
 ### Disagreements Identified
 
 **Disagreement with Judge {X} on Criterion "{Name}"**
+
 - My score: {my_score}/5
 - Their score: {their_score}/5
 - My defense: [quote evidence supporting my score]
@@ -192,6 +196,7 @@ Instructions:
 ### Revised Assessment
 
 After considering other judges' arguments:
+
 - **Criterion "{Name}"**: [Maintained {X}/5 | Revised from {X} to {Y}/5]
   - Reason for change: [what convinced me] OR
   - Reason maintained: [why I stand by original score]
@@ -201,11 +206,13 @@ After considering other judges' arguments:
 **New Weighted Score**: {updated_total}/5.0
 
 ## Evidences
+
 [specific quotes]
 
---- 
+---
 
 CRITICAL:
+
 - Only revise if you find their evidence compelling
 - Defend your original scores if you still believe them
 - Quote specific evidence from the solution
@@ -216,11 +223,13 @@ CRITICAL:
 After each debate round, check for consensus:
 
 **Consensus achieved if:**
+
 - All judges' overall scores within 0.5 points of each other
 - No criterion has >1 point disagreement across any two judges
 - All judges explicitly state they accept the consensus
 
 **If no consensus after 3 rounds:**
+
 - Report persistent disagreements
 - Provide all judge reports for human review
 - Flag that automated evaluation couldn't reach consensus
@@ -230,7 +239,7 @@ After each debate round, check for consensus:
 **Step 1: Run Independent Analysis (Round 1)**
 
 1. Launch 3 judge agents in parallel (Judge 1, 2, 3)
-2. Each writes their independent assessment to `.specs/reports/{solution-name}-{date}.[1|2|3].md`
+2. Each writes their independent assessment to `.opencode/reports/{solution-name}-{date}.[1|2|3].md`
 3. Wait for all 3 agents to complete
 
 **Step 2: Check for Consensus**
@@ -238,10 +247,12 @@ After each debate round, check for consensus:
 Let's work through this systematically to ensure accurate consensus detection.
 
 Read all three reports and extract:
+
 - Each judge's overall weighted score
 - Each judge's score for every criterion
 
 Check consensus step by step:
+
 1. First, extract all overall scores from each report and list them explicitly
 2. Calculate the difference between the highest and lowest overall scores
    - If difference ≤ 0.5 points → overall consensus achieved
@@ -288,10 +299,10 @@ Let's synthesize the evaluation results step by step.
      - Number of rounds to reach consensus
      - Final recommendation with clear justification
    - If there is no consensus:
-       - All judges' final scores showing disagreements
-       - Specific criteria where consensus wasn't reached
-       - Analysis of why consensus couldn't be reached
-       - Flag for human review
+     - All judges' final scores showing disagreements
+     - Specific criteria where consensus wasn't reached
+     - Analysis of why consensus couldn't be reached
+     - Flag for human review
 4. Command complete
 
 ### Phase 3: Consensus Report
@@ -308,25 +319,31 @@ Let's compile the final consensus by analyzing each component systematically.
 First, let's consolidate all judges' final scores:
 
 | Criterion | Judge 1 | Judge 2 | Judge 3 | Final |
-|-----------|---------|---------|---------|-------|
+| --------- | ------- | ------- | ------- | ----- |
 | {Name}    | {X}/5   | {X}/5   | {X}/5   | {X}/5 |
+
 ...
 
 **Consensus Overall Score**: {avg}/5.0
 
 ## Consensus Strengths
+
 [Review each judge's identified strengths and extract the common themes that all judges agreed upon]
 
 ## Consensus Weaknesses
+
 [Review each judge's identified weaknesses and extract the common themes that all judges agreed upon]
 
 ## Debate Summary
+
 Let's trace how consensus was reached:
+
 - Rounds to consensus: {N}
 - Initial disagreements: {list with specific criteria and score gaps}
 - How resolved: {for each disagreement, explain what evidence or argument led to resolution}
 
 ## Final Recommendation
+
 Based on the consensus scores and the key strengths/weaknesses identified:
 {Pass/Fail/Needs Revision with clear justification tied to the evidence}
 ```
@@ -334,11 +351,11 @@ Based on the consensus scores and the key strengths/weaknesses identified:
 <output>
 The command produces:
 
-1. **Reports directory**: `.specs/reports/` (created if not exists)
-2. **Initial reports**: `.specs/reports/{solution-name}-{date}.1.md`, `.specs/reports/{solution-name}-{date}.2.md`, `.specs/reports/{solution-name}-{date}.3.md`
+1. **Reports directory**: `.opencode/reports/` (created if not exists)
+2. **Initial reports**: `.opencode/reports/{solution-name}-{date}.1.md`, `.opencode/reports/{solution-name}-{date}.2.md`, `.opencode/reports/{solution-name}-{date}.3.md`
 3. **Debate updates**: Appended sections in each report file per round
 4. **Final synthesis**: Replied to user (consensus or disagreement summary)
-</output>
+   </output>
 
 ## Best Practices
 
@@ -347,6 +364,7 @@ The command produces:
 Choose 3-5 weighted criteria relevant to the solution type:
 
 **Code evaluation:**
+
 - Correctness (30%) - Does it work? Handles edge cases?
 - Design Quality (25%) - Clean architecture? Maintainable?
 - Efficiency (20%) - Performance considerations?
@@ -354,6 +372,7 @@ Choose 3-5 weighted criteria relevant to the solution type:
 - Testing (10%) - Test coverage? Test quality?
 
 **Design/Architecture evaluation:**
+
 - Completeness (30%) - All requirements addressed?
 - Feasibility (25%) - Can it actually be built?
 - Scalability (20%) - Handles growth?
@@ -361,6 +380,7 @@ Choose 3-5 weighted criteria relevant to the solution type:
 - Documentation (10%) - Clear and comprehensive?
 
 **Documentation evaluation:**
+
 - Accuracy (35%) - Technically correct?
 - Completeness (30%) - Covers all necessary topics?
 - Clarity (20%) - Easy to understand?
@@ -392,28 +412,33 @@ Choose 3-5 weighted criteria relevant to the solution type:
 ```
 
 **Round 1 outputs** (assuming date 2025-01-15):
-- `.specs/reports/users-api-2025-01-15.1.md` - Judge 1 scores correctness 4/5, security 3/5
-- `.specs/reports/users-api-2025-01-15.2.md` - Judge 2 scores correctness 4/5, security 5/5
-- `.specs/reports/users-api-2025-01-15.3.md` - Judge 3 scores correctness 5/5, security 4/5
+
+- `.opencode/reports/users-api-2025-01-15.1.md` - Judge 1 scores correctness 4/5, security 3/5
+- `.opencode/reports/users-api-2025-01-15.2.md` - Judge 2 scores correctness 4/5, security 5/5
+- `.opencode/reports/users-api-2025-01-15.3.md` - Judge 3 scores correctness 5/5, security 4/5
 
 **Disagreement detected:** Security scores range from 3-5
 
 **Round 2 debate:**
+
 - Judge 1 defends 3/5: "Missing rate limiting, input validation incomplete"
 - Judge 2 challenges: "Rate limiting exists in middleware (line 45)"
 - Judge 1 revises to 4/5: "Missed middleware, but input validation still weak"
 - Judge 3 defends 4/5: "Input validation adequate for requirements"
 
 **Round 2 outputs:**
+
 - All judges now 4-5/5 on security (within 1 point)
 - Disagreement on input validation remains
 
 **Round 3 debate:**
+
 - Judges examine specific validation code
 - Judge 2 revises to 4/5: "Upon re-examination, email validation regex is weak"
 - Consensus: Security = 4/5
 
 **Final consensus:**
+
 ```
 Correctness: 4.3/5
 Design: 4.5/5
@@ -423,4 +448,3 @@ Documentation: 4.0/5
 
 Overall: 4.3/5 - PASS
 ```
-

package/skills/sadd-judge-with-debate/manifest.json

@@ -9,9 +9,7 @@
     "consensus judge",
     "llm judge"
   ],
-  "applicable_agents": [
-    "critic"
-  ],
+  "applicable_agents": ["planner"],
   "max_context_tokens": 2200,
   "entry_file": "SKILL.md"
 }

package/skills/structured-code-review/manifest.json

@@ -2,17 +2,8 @@
   "name": "structured-code-review",
   "version": "1.0.0",
   "description": "Structured checklist for local code review passes",
-  "triggers": [
-    "review",
-    "code review",
-    "check code",
-    "examine"
-  ],
-  "applicable_agents": [
-    "reviewer",
-    "worker",
-    "heavy-worker"
-  ],
+  "triggers": ["review", "code review", "check code", "examine"],
+  "applicable_agents": ["planner", "reviewer", "auditor"],
   "max_context_tokens": 2000,
   "entry_file": "SKILL.md"
 }

package/skills/task-decomposition/manifest.json

@@ -2,19 +2,8 @@
   "name": "task-decomposition",
   "version": "1.0.0",
   "description": "Break larger requests into clear executable subtasks",
-  "triggers": [
-    "decompose",
-    "break down",
-    "split",
-    "tasks",
-    "planning",
-    "subtasks"
-  ],
-  "applicable_agents": [
-    "planner",
-    "executor",
-    "worker"
-  ],
+  "triggers": ["decompose", "break down", "split", "tasks", "planning", "subtasks"],
+  "applicable_agents": ["planner"],
   "max_context_tokens": 1500,
   "entry_file": "SKILL.md"
 }

package/skills/verification-before-completion/manifest.json

@@ -2,21 +2,8 @@
   "name": "verification-before-completion",
   "version": "1.0.0",
   "description": "Verification checklist before claiming work complete",
-  "triggers": [
-    "verify",
-    "complete",
-    "done",
-    "finish",
-    "check",
-    "validation"
-  ],
-  "applicable_agents": [
-    "worker",
-    "heavy-worker",
-    "deep-worker",
-    "ui-worker",
-    "ui-heavy-worker"
-  ],
+  "triggers": ["verify", "complete", "done", "finish", "check", "validation"],
+  "applicable_agents": ["coder", "ui-coder", "sec-coder"],
   "max_context_tokens": 2000,
   "entry_file": "SKILL.md"
 }

package/skills/verification-gates/SKILL.md

@@ -16,9 +16,8 @@ tags:
 difficulty: beginner
 estimatedTime: 10
 relatedSkills:
-  - planning/design-first
-  - planning/task-decomposition
-  - testing/red-green-refactor
+  - design-first
+  - task-decomposition
 ---
 
 # Verification Gates
@@ -44,6 +43,7 @@ Before starting design:
 - [ ] Edge cases are identified
 
 **Actions:**
+
 1. Review requirements document
 2. Identify any unclear items
 3. Get explicit confirmation on ambiguous points
@@ -61,6 +61,7 @@ Before starting implementation:
 - [ ] Design has been reviewed (self or peer)
 
 **Actions:**
+
 1. Walk through design against requirements
 2. Review with rubber duck or teammate
 3. Check for missing pieces
@@ -78,6 +79,7 @@ Before calling task complete:
 - [ ] Dependencies are appropriate
 
 **Actions:**
+
 1. Run full test suite
 2. Self-review the diff
 3. Check for code smells
@@ -94,6 +96,7 @@ Before merging:
 - [ ] Documentation is updated
 
 **Actions:**
+
 1. Test the full user flow
 2. Run integration test suite
 3. Compare performance metrics
@@ -110,6 +113,7 @@ Before marking complete:
 - [ ] Stakeholders can verify
 
 **Actions:**
+
 1. Smoke test in environment
 2. Check error logs and metrics
 3. Get stakeholder sign-off
@@ -150,6 +154,7 @@ Gates requiring human judgment:
 ## Manual Verification Checklist
 
 Before Code Review:
+
 - [ ] I've tested my changes locally
 - [ ] I've written/updated tests
 - [ ] I've read my own diff
@@ -157,6 +162,7 @@ Before Code Review:
 - [ ] I've updated documentation
 
 Before Deployment:
+
 - [ ] Code review approved
 - [ ] QA verified (if applicable)
 - [ ] Stakeholder approved (if required)
@@ -167,13 +173,13 @@ Before Deployment:
 
 Gates that apply in specific situations:
 
-| Condition | Required Gates |
-|-----------|---------------|
-| Security-related | Security review |
-| Public API change | API review + migration plan |
-| Database change | DBA review + backup plan |
-| Performance-sensitive | Performance test |
-| Breaking change | Deprecation notice + migration |
+| Condition             | Required Gates                 |
+| --------------------- | ------------------------------ |
+| Security-related      | Security review                |
+| Public API change     | API review + migration plan    |
+| Database change       | DBA review + backup plan       |
+| Performance-sensitive | Performance test               |
+| Breaking change       | Deprecation notice + migration |
 
 ## Implementing Gates
 
@@ -213,14 +219,17 @@ Task Complete
 **Purpose:** [What this gate ensures]
 
 **Checklist:**
+
 - [ ] Item 1
 - [ ] Item 2
 - [ ] Item 3
 
 **Verification Method:**
+
 - [How to verify each item]
 
 **Failure Actions:**
+
 - [What to do if gate fails]
 
 **Approver:** [Who can approve passage]
@@ -265,17 +274,16 @@ jobs:
 
 ## Quick Reference
 
-| Phase | Gate Before | Key Checks |
-|-------|-------------|------------|
-| Design | Requirements | Clear, complete, approved |
-| Implementation | Design | Reviewed, feasible |
-| Review | Implementation | Tests, conventions, working |
-| Merge | Review | Approved, conflicts resolved |
-| Deploy | Merge | Environment ready, plan exists |
+| Phase          | Gate Before    | Key Checks                     |
+| -------------- | -------------- | ------------------------------ |
+| Design         | Requirements   | Clear, complete, approved      |
+| Implementation | Design         | Reviewed, feasible             |
+| Review         | Implementation | Tests, conventions, working    |
+| Merge          | Review         | Approved, conflicts resolved   |
+| Deploy         | Merge          | Environment ready, plan exists |
 
 ## Integration with Other Skills
 
 - **design-first**: Gates validate design before implementation
 - **task-decomposition**: Gates between task phases
-- **testing/red-green-refactor**: Tests are key gate criteria
-- **collaboration/structured-review**: Review is a gate
+- **structured-code-review**: Review is a gate

package/skills/verification-gates/manifest.json

@@ -2,18 +2,8 @@
   "name": "verification-gates",
   "version": "1.0.0",
   "description": "Verification gates and acceptance checkpoints between risky phases",
-  "triggers": [
-    "verification gate",
-    "quality gate",
-    "milestone check",
-    "approval step",
-    "go no go"
-  ],
-  "applicable_agents": [
-    "planner",
-    "auditor",
-    "strategist"
-  ],
+  "triggers": ["verification gate", "quality gate", "milestone check", "approval step", "go no go"],
+  "applicable_agents": ["planner", "auditor"],
   "max_context_tokens": 1800,
   "entry_file": "SKILL.md"
 }

package/agents/advisor.md

@@ -1,57 +0,0 @@
----
-description: Read-only coding advisor for workers that are blocked, low-confidence, or choosing between implementation paths
-mode: subagent
-model: openai/gpt-5.4
-temperature: 0
-steps: 20
-permission:
-  "*": deny
-  read:
-    "*": allow
-    "*.env": deny
-    "*.env.*": deny
-    "*.env.example": allow
-  glob: allow
-  grep: allow
-  list: allow
-  lsp: allow
-  todoread: allow
-  todowrite: allow
-  code_index_set_project_path: allow
-  code_index_search_code_advanced: allow
-  code_index_find_files: allow
-  code_index_get_file_summary: allow
-  code_index_get_symbol_body: allow
-  repo_git_status: allow
-  repo_git_diff: allow
-  task: deny
-  skill:
-    "*": deny
-    task-decomposition: allow
-    root-cause-analysis: allow
-    design-first: allow
-  edit: deny
-  bash: allow
-  webfetch: deny
-  websearch: deny
-  codesearch: deny
-  external_directory: allow
----
-
-You are `advisor`.
-
-Role
-- Help workers when they are stuck, low-confidence, or choosing between implementation paths.
-- Inspect local code and return direction without editing.
-- Prefer the narrowest answer that unblocks the worker's assigned task.
-
-Caller contract
-- Provide the task, the blocker, and the most relevant files or symbols.
-- Ask one focused question.
-
-Output
-- `## Bottom Line`
-- `## Diagnosis`
-- `## Recommended Next Edits`
-- `## Checks`
-- `## Uncertainty`