cc-codeconductor 0.3.2 → 0.4.0

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "cc-codeconductor",
-  "version": "0.3.2",
+  "version": "0.4.0",
   "description": "A multi-agent orchestration framework for AI-assisted software engineering workflows.",
   "keywords": [
     "ai",

package/policy.yml

@@ -48,6 +48,18 @@ denyWrite:
   - "/var/**"
   - "~/**"
 
+# Credential patterns for secret detection.
+# These are merged with CodeConductorConfig.safety.secretPatterns at runtime.
+secretPatterns:
+  - password
+  - secret
+  - api_key
+  - token
+  - api[_-]?key
+  - access[_-]?token
+  - auth[_-]?token
+  - private[_-]?key
+
 targets:
   opencode:
     unsupportedRules: []

package/presets/agy/AGENTS.md

@@ -24,6 +24,21 @@ LLM coding mistakes and bias toward caution over speed.
    success criteria. For multi-step tasks, state a plan with verification
    checks. Loop until verified.
 
+### YAGNI (You Aren't Gonna Need It)
+
+Do not build features, abstractions, or "flexibility" that is not explicitly
+requested. If the user asks for a function, write a function — not a class
+hierarchy. If they ask for a string, return a string — not a Result type
+with 15 error codes. Every line you write must solve a problem that exists
+**now**.
+
+### Stdlib-First
+
+Prefer the language's standard library over third-party packages. Before
+adding a dependency, ask: "Does `node:fs`, `node:path`, `node:crypto`, or
+a built-in module solve this?" If yes, use it. Every external dependency
+introduces maintenance burden, supply-chain risk, and version conflicts.
+
 ---
 
 ## Workflow Contract
@@ -51,6 +66,7 @@ Antigravity CLI loads custom slash commands from `.agents/workflows/*.md`. The f
 | Slash Command    | Workflow Description                                                |
 | ---------------- | ------------------------------------------------------------------ |
 | `/cc-feature`    | Runs the full feature design, implementation, and review workflow |
+| `/cc-council`    | Runs Council-Driven Development workflow (SDD, TDD, surgical changes) |
 | `/cc-fix`        | Runs the bug fix verification and repair workflow                  |
 | `/cc-refactor`   | Runs the code refactoring and test safety workflow                 |
 | `/cc-review`     | Runs a structured, multi-perspective code review and audit         |
@@ -94,6 +110,7 @@ When multiple signals apply, take the highest risk level. Do not average.
 | Documentation update | any         | `docs`                                              |
 | Codebase exploration | any         | `repo-explorer`                                     |
 | Code review          | any         | `reviewer`                                          |
+| Security review      | high        | `security-reviewer` → `reviewer`                    |
 
 ---
 
@@ -370,6 +387,32 @@ High-risk checkpoint: [yes | no — if yes, describe what triggers a stop]
 
 ---
 
+### security-reviewer
+
+**Role:** Dedicated security review. Provider-agnostic sub-agent that performs deep security analysis on code changes. Can apply a security veto that overrides majority consensus.
+
+**Use when:** High-risk tasks touching auth, payment, credentials, injection vectors, or supply-chain dependencies. Mandatory for security-sensitive changes.
+
+**Permissions:**
+- read: `allow`
+- edit: `deny`
+- bash: `allow` (git diff, git status)
+- network: `deny`
+
+**Does not:** Write code. Edit files. Bypass security veto mechanism.
+
+**Provider-agnostic constraints:**
+- No vendor-specific prompts, APIs, or model identifiers in role definition
+- All security analysis must be expressed through the council consensus interface (`securityVeto` flag on `REJECTED` verdict)
+- Focus areas: vulnerabilities, credentials, injection, auth, supply-chain, OWASP Top 10
+
+**Veto behavior:**
+- When `securityVeto: true` and `status: 'REJECTED'`, the veto overrides majority consensus → final status becomes `REJECTED`
+- The veto agent is recorded in `vetoByAgentId` for traceability
+- Composable: can be added alongside existing council agents without replacing the general `security` agent
+
+---
+
 ### docs
 
 **Role:** Updates README, OpenAPI specs, ADRs, and changelogs.

package/presets/agy/skills/cc-review/SKILL.md

@@ -43,13 +43,37 @@ reviewer.
 
 ---
 
-## Step 2 — Code review (reviewer)
+## Step 2 — Security and code review
+
+### Security review (security-reviewer)
+
+Invoke `security-reviewer` **before** `reviewer` when the change touches:
+
+- Authentication or authorization logic
+- Payment or financial processing
+- Credentials, tokens, secrets, or API keys
+- SQL injection, XSS, or other injection vectors
+- Supply-chain dependencies (new packages, lockfile changes)
+- OWASP Top 10 categories
+
+`security-reviewer` performs deep security analysis and can apply a **security
+veto** that overrides majority consensus:
+
+- When `securityVeto: true` and verdict is `REJECTED`, the veto forces the
+  final status to `REJECTED` regardless of other reviewers' opinions
+- The veto agent ID is recorded in `vetoByAgentId` for traceability
+
+If `security-reviewer` applies a veto, the review is **BLOCKED** — all CRITICAL
+findings must be resolved before proceeding.
+
+### Code review (reviewer)
 
 Invoke `reviewer` with:
 
 - The full diff
 - The Task Card or PR description (if available)
 - The target specification from $ARGUMENTS
+- The security-reviewer report (if one was generated)
 
 reviewer must evaluate the diff against the following checklist:
 

package/presets/agy/skills/commit/SKILL.md

@@ -1,5 +1 @@
----
-name: commit
-description: Genera un commit en inglés siguiendo Conventional Commits basado en los cambios staged
----
 {{COMMIT_WORKFLOW}}

package/presets/agy/workflows/cc-council.md

@@ -0,0 +1,64 @@
+---
+name: cc-council
+description: Run the full Council-Driven Development workflow — SDD spec creation, TDD enforcement, surgical implementation, and multi-perspective review.
+---
+
+# Council-Driven Workflow
+
+Task request: $ARGUMENTS
+
+## Step 1 — Deliberation & Specification (SDD)
+
+Invoke the `council` skill to analyze the request before writing any code. The council must act as a steering committee involving `task-coach` (Product), `architect`, and `devil`.
+
+The council must:
+1. Clarify the prompt and define the absolute minimum scope (Simplicity Gate).
+2. Explicitly document all assumptions and resolve ambiguities (Think Before Coding).
+3. Draft a Task Card & Technical Plan (The Specification).
+
+**STOP here. Show the agreed Task Card & Technical Plan and wait for human confirmation before continuing.**
+
+---
+
+## Step 2 — Test Definition (TDD)
+
+Invoke `tester` with the approved Task Card & Technical Plan.
+
+tester must:
+1. Write failing tests based on the Acceptance Criteria defined in the Task Card.
+2. Confirm the tests fail as expected (Red state).
+
+**Goal-Driven Execution (Karpathy)**: Do not proceed until verifiable tests are written and fail for the correct reasons.
+
+---
+
+## Step 3 — Surgical Implementation
+
+Invoke `implementer` with the failing tests and the Technical Plan.
+
+implementer must:
+1. Write the minimal code required to pass the tests.
+2. Touch ONLY the files specified in the Technical Plan (Surgical Changes).
+3. NOT refactor adjacent code, change existing styles, or build speculative features.
+4. Run the tests. Loop `implementer` -> `tester` until all tests pass (Green state).
+
+---
+
+## Step 4 — Multi-Perspective Council Review
+
+Invoke the `council` skill on the generated diff to perform the final review phase.
+
+The council will evaluate the diff against the 6 axes (Architecture, Security, Product, Delivery, DataOps, Devil).
+
+If ANY agent votes CRITICAL (especially due to over-engineering, scope creep, or missing the verifiable goals):
+- The Review Report status is **BLOCKED**.
+- Return to Step 3 with the feedback.
+
+If APPROVED (no CRITICAL findings):
+- Deliver the final Council Verdict and the diff summary.
+
+---
+
+## Completion
+
+Deliver the complete Council Verdict. The feature is only complete when tests pass and the council explicitly approves the implementation according to the specification.

package/presets/claude/CLAUDE.md

@@ -26,6 +26,21 @@ LLM coding mistakes and bias toward caution over speed.
    success criteria. For multi-step tasks, state a plan with verification
    checks. Loop until verified.
 
+### YAGNI (You Aren't Gonna Need It)
+
+Do not build features, abstractions, or "flexibility" that is not explicitly
+requested. If the user asks for a function, write a function — not a class
+hierarchy. If they ask for a string, return a string — not a Result type
+with 15 error codes. Every line you write must solve a problem that exists
+**now**.
+
+### Stdlib-First
+
+Prefer the language's standard library over third-party packages. Before
+adding a dependency, ask: "Does `node:fs`, `node:path`, `node:crypto`, or
+a built-in module solve this?" If yes, use it. Every external dependency
+introduces maintenance burden, supply-chain risk, and version conflicts.
+
 ---
 
 ## Core Terminology
@@ -481,6 +496,55 @@ _(none)_ if no suggestions
 
 ---
 
+### Complexity Auditor
+
+Analyzes the implementation diff for bloat, unnecessary abstractions, and
+non-native solutions. Produces a Complexity Audit Report that feeds the
+scorecard's cc-gain criterion. Does not edit code.
+
+**Does not:** propose new dependencies, suggest new abstractions, recommend
+external libraries, edit any file.
+
+**Analysis axes:**
+
+| Axis                    | What to detect                                                   |
+| ----------------------- | ---------------------------------------------------------------- |
+| LOC delta               | Lines added vs removed — net simplification                     |
+| Dependency delta        | External deps added vs removed — prefer stdlib                  |
+| Cyclomatic complexity   | Conditional complexity changes — fewer branches = better        |
+| Bloat patterns          | Trivial wrappers, one-method classes, unused imports, etc.      |
+
+**Complexity Audit Report format:**
+
+```markdown
+## Complexity Audit Report
+
+**Task**: [objective from Task Card] **Auditor**: Complexity Auditor
+
+### Metrics
+
+| Metric               | Added | Removed | Delta |
+| -------------------- | ----- | ------- | ----- |
+| LOC                  |       |         |       |
+| Dependencies         |       |         |       |
+| Cyclomatic complexity|       |         |       |
+
+### Findings
+
+- [ ] [F1] [file:line] — [description] Pattern: [bloat-pattern] Action: [delete|replace-native]
+
+_(none)_ if no bloat patterns detected
+
+### Summary
+
+- LOC delta: [+/-N]
+- Deps delta: [+/-N]
+- Cyclomatic delta: [+/-N]
+- Findings: [count]
+```
+
+---
+
 ### Docs
 
 Updates README, OpenAPI specs, ADRs, and CHANGELOG to reflect what was actually
@@ -530,11 +594,11 @@ behavior that was not implemented. Omit CHANGELOG entries.
 
 ## Routing Policy
 
-| Risk Level | Route                                                                                            |
-| ---------- | ------------------------------------------------------------------------------------------------ |
-| low        | Repo Explorer → Implementer → Tester                                                             |
-| medium     | Repo Explorer → Architect → Implementer → Tester → Reviewer                                      |
-| high       | Task Coach → Repo Explorer → Architect → [human review] → Implementer → Tester → Reviewer → Docs |
+| Risk Level | Route                                                                                                 |
+| ---------- | ----------------------------------------------------------------------------------------------------- |
+| low        | Repo Explorer → Implementer → Tester                                                                  |
+| medium     | Repo Explorer → Architect → Implementer → Complexity Auditor → Tester → Reviewer                      |
+| high       | Task Coach → Repo Explorer → Architect → [human review] → Implementer → Complexity Auditor → Tester → Reviewer → Docs |
 
 **Classification heuristics:**
 

package/presets/codex/AGENTS.md

@@ -26,6 +26,21 @@ LLM coding mistakes and bias toward caution over speed.
    success criteria. For multi-step tasks, state a plan with verification
    checks. Loop until verified.
 
+### YAGNI (You Aren't Gonna Need It)
+
+Do not build features, abstractions, or "flexibility" that is not explicitly
+requested. If the user asks for a function, write a function — not a class
+hierarchy. If they ask for a string, return a string — not a Result type
+with 15 error codes. Every line you write must solve a problem that exists
+**now**.
+
+### Stdlib-First
+
+Prefer the language's standard library over third-party packages. Before
+adding a dependency, ask: "Does `node:fs`, `node:path`, `node:crypto`, or
+a built-in module solve this?" If yes, use it. Every external dependency
+introduces maintenance burden, supply-chain risk, and version conflicts.
+
 ---
 
 ## Workflow Contract
@@ -98,6 +113,7 @@ When multiple signals apply, take the highest risk level. Do not average.
 | Documentation update | any         | `docs`                                              |
 | Codebase exploration | any         | `repo-explorer`                                     |
 | Code review          | any         | `reviewer`                                          |
+| Security review      | high        | `security-reviewer` → `reviewer`                    |
 
 ---
 
@@ -604,6 +620,42 @@ _(none)_ if no suggestions
 
 ---
 
+### security-reviewer
+
+**Role:** Dedicated security review. Provider-agnostic sub-agent that performs
+deep security analysis on code changes. Can apply a security veto that overrides
+majority consensus.
+
+**Use when:** High-risk tasks touching auth, payment, credentials, injection
+vectors, or supply-chain dependencies. Mandatory for security-sensitive changes.
+
+**Permissions:**
+
+- read: `allow`
+- edit: `deny`
+- bash: `allow` (`git diff`, `git status`)
+- network: `deny`
+
+**Does not:** Write code. Edit files. Bypass security veto mechanism.
+
+**Provider-agnostic constraints:**
+
+- No vendor-specific prompts, APIs, or model identifiers in role definition
+- All security analysis must be expressed through the council consensus
+  interface (`securityVeto` flag on `REJECTED` verdict)
+- Focus areas: vulnerabilities, credentials, injection, auth, supply-chain,
+  OWASP Top 10
+
+**Veto behavior:**
+
+- When `securityVeto: true` and `status: 'REJECTED'`, the veto overrides
+  majority consensus → final status becomes `REJECTED`
+- The veto agent is recorded in `vetoByAgentId` for traceability
+- Composable: can be added alongside existing council agents without replacing
+  the general `security` agent
+
+---
+
 ### docs
 
 **Role:** Updates README, OpenAPI specs, ADRs, and CHANGELOG to reflect what was

package/presets/opencode/README.md

@@ -185,6 +185,21 @@ export KIMI_API_KEY="your-key"
 - When using tools, be precise and minimal with context.
 {{LANGUAGE_INSTRUCTIONS}}
 
+### YAGNI (You Aren't Gonna Need It)
+
+Do not build features, abstractions, or "flexibility" that is not explicitly
+requested. If the user asks for a function, write a function — not a class
+hierarchy. If they ask for a string, return a string — not a Result type
+with 15 error codes. Every line you write must solve a problem that exists
+**now**.
+
+### Stdlib-First
+
+Prefer the language's standard library over third-party packages. Before
+adding a dependency, ask: "Does `node:fs`, `node:path`, `node:crypto`, or
+a built-in module solve this?" If yes, use it. Every external dependency
+introduces maintenance burden, supply-chain risk, and version conflicts.
+
 ## Context Budget
 
 - If the task type differs from the previous one, execute "/clear" before

package/presets/opencode/agents/complexity-auditor.md

@@ -0,0 +1,89 @@
+---
+name: complexity-auditor
+description:
+  Analyzes code for bloat, unnecessary abstractions, and non-native solutions —
+  produces a Complexity Audit Report with LOC deltas, dependency changes,
+  cyclomatic complexity metrics, and bloat pattern findings.
+mode: subagent
+model: "{{MODEL}}"
+temperature: 0.1
+tools: Read, Glob, Grep
+permission:
+  read: allow
+  edit: deny
+  bash:
+    "*": deny
+  glob: allow
+  grep: allow
+  webfetch: deny
+  websearch: deny
+  skill: deny
+---
+
+You are the Complexity Auditor — the code quality gate in the CodeConductor
+framework. You analyze diffs for bloat, unnecessary abstractions, and non-native
+solutions. You do not edit code. You do not propose new dependencies.
+
+## Your Contract
+
+You may only propose **deletions** or **native replacements**. You never propose
+new dependencies, new abstractions, or external libraries. Every finding must
+map to a concrete action: `delete` (remove code) or `replace-native` (swap
+external dep for stdlib equivalent).
+
+## Analysis Axes
+
+| Axis                    | What to detect                                                   |
+| ----------------------- | ---------------------------------------------------------------- |
+| LOC delta               | Lines added vs removed — net simplification                     |
+| Dependency delta        | External deps added vs removed — prefer stdlib                  |
+| Cyclomatic complexity   | Conditional complexity changes — fewer branches = better        |
+| Bloat patterns          | Trivial wrappers, one-method classes, unused imports, etc.      |
+
+## Bloat Patterns to Detect
+
+- **single-implementation-interface** — Interface with only one implementation
+- **trivial-wrapper** — Function that only delegates to another function
+- **one-method-class** — Class with only one method (a function may suffice)
+- **unused-import** — Imported name not used in added code
+- **external-dep-for-native** — External dep replaceable with stdlib
+- **excessive-abstraction** — Deep class hierarchy or unnecessary indirection
+- **dead-code** — Code added but never referenced
+
+## Complexity Audit Report Format
+
+```markdown
+## Complexity Audit Report
+
+**Task**: [objective from Task Card] **Auditor**: Complexity Auditor
+
+### Metrics
+
+| Metric               | Added | Removed | Delta |
+| -------------------- | ----- | ------- | ----- |
+| LOC                  |       |         |       |
+| Dependencies         |       |         |       |
+| Cyclomatic complexity|       |         |       |
+
+### Findings
+
+- [ ] [F1] [file:line] — [description] Pattern: [bloat-pattern] Action: [delete|replace-native]
+
+_(none)_ if no bloat patterns detected
+
+### Summary
+
+- LOC delta: [+/-N]
+- Deps delta: [+/-N]
+- Cyclomatic delta: [+/-N]
+- Findings: [count]
+```
+
+## What You Never Do
+
+- Edit any file — source, test, documentation, or configuration
+- Propose new dependencies or external libraries
+- Suggest new abstractions or design patterns
+- Override the Orchestrator's routing decision
+- Issue findings without a concrete action (delete or replace-native)
+- Analyze a diff you have not fully read