By sc30gsw
Verified Coherence Spec-Driven Development — adversarial quality gates for AI-assisted development
npx claudepluginhub sc30gsw/vcsdd-claude-code --plugin vcsddRun Phase 3 (adversarial review) for the active VCSDD feature. Spawns a fresh vcsdd-adversary agent with zero Builder context to review implementation against spec. Produces binary PASS/FAIL verdict per dimension.
Run BFS forward-impact analysis from changed nodes through the CEG.
Rebuild the Conditioned Evidence Graph (CEG) from `coherence:` or `codd:`
Validate the CEG for reference integrity and circular dependencies.
Create a git commit for the current VCSDD phase. Auto-generates a conventional commit message from state.json including phase, sprint, gate verdict, and affected bead IDs. Tags with vcsdd/<feature>/phase-<id>.
Run the strict-mode sprint contract review after Phase 2c. Spawns a fresh vcsdd-adversary instance to review `contracts/sprint-N.md` and writes a PASS/FAIL verdict under `reviews/contracts/sprint-N/output/`.
Run Phase 6 (convergence check) for the active VCSDD feature. Validates four-dimensional convergence: finding diminishment, finding specificity, criteria coverage, and duplicate detection.
Architect escalation approval — extend iteration limit for a phase that has hit its cap. Records the approval decision in state and allows the blocked transition to proceed.
Run Phase 4 (feedback routing) after an adversarial review FAIL. Routes adversary findings to the earliest affected phase, checks iteration limits, and escalates to human if limits are exceeded.
Run Phase 5 (formal hardening) for the active VCSDD feature. Invokes vcsdd-verifier to execute proof obligations, security hardening, and a purity-boundary audit before convergence.
Run Phase 2b (implementation, Green phase) and Phase 2c (refactor) for the active VCSDD feature. Invokes vcsdd-builder to implement minimal code to pass failing tests, then refactor.
Initialize a VCSDD feature pipeline. Creates the .vcsdd/features/<name>/ directory tree, sets mode (strict|lean), and activates the feature. Must be run before any other VCSDD command.
Run Phase 1c (spec review gate) for the active VCSDD feature. Spawns a fresh vcsdd-adversary instance to review the behavioral spec and verification architecture. Records PASS/FAIL gate verdict.
Run Phase 1a (behavioral specification) and Phase 1b (verification architecture) for the active VCSDD feature. Invokes vcsdd-builder to write EARS-format requirements and purity boundary analysis.
Display the current VCSDD pipeline status for the active (or specified) feature. Shows phase, mode, sprint count, open findings, pending proofs, and the traceability summary.
Run Phase 2a (test generation, Red phase) for the active VCSDD feature. Invokes vcsdd-builder to generate failing tests for all spec requirements. Records red phase evidence.
Display the full traceability chain for a VCSDD bead. Traverses the Chainlink graph from the given bead ID and shows all connected artifacts with status.
VCSDD adversarial reviewer with fresh context. Use this agent ONLY for Phase 1c (spec review), strict-mode contract review, and Phase 3 (implementation review). This agent must be spawned as a NEW instance with zero Builder context. It reviews artifacts from disk and writes verdict/findings only under the feature review output directory. It produces binary PASS/FAIL verdicts per dimension with concrete findings.
VCSDD spec author and TDD implementer. Use this agent to write behavioral specifications, generate failing tests (Red phase), implement code to pass tests (Green phase), refactor code, write sprint contracts, and create verification architectures. Invoked for all code-writing and spec-writing tasks in the VCSDD pipeline.
VCSDD pipeline state manager and gate enforcer. Use this agent to coordinate the overall VCSDD workflow, track phase transitions, manage sprint contracts, and enforce quality gates. Invoke when you need to advance the pipeline, check gate prerequisites, route adversary feedback, or manage the .vcsdd/ state directory.
VCSDD formal verification coordinator. Use this agent for Phase 5 (Formal Hardening). It reads the language profile, detects installed verification tools, writes proof harnesses, runs proof/security/purity checks, and produces verification-report.md, security-report.md, and purity-audit.md. Invoked by /vcsdd-harden command.
Use this skill when running or interpreting VCSDD adversarial reviews. Provides calibration guidance for the Adversary agent, finding severity classification, and anti-leniency enforcement patterns.
Run Phase 3 (adversarial review) for the active VCSDD feature. Spawns a fresh vcsdd-adversary agent with zero Builder context to review implementation against spec. Produces binary PASS/FAIL verdict per dimension.
Given a set of changed or affected spec nodes, perform BFS forward-impact
Rebuild the Conditioned Evidence Graph (CEG) from `coherence:` or `codd:`
Validate the CEG for structural integrity:
Create a git commit for the current VCSDD phase. Auto-generates a conventional commit message from state.json including phase, sprint, gate verdict, and affected bead IDs. Tags with vcsdd/<feature>/phase-<id>.
Run the strict-mode sprint contract review after Phase 2c. Spawns a fresh vcsdd-adversary instance to review `contracts/sprint-N.md` and writes a PASS/FAIL verdict under `reviews/contracts/sprint-N/output/`.
Run Phase 6 (convergence check) for the active VCSDD feature. Validates four-dimensional convergence: finding diminishment, finding specificity, criteria coverage, and duplicate detection.
Use this skill during Phase 6 convergence checks. Provides four-dimensional convergence analysis, hallucination detection, and duplicate finding identification.
Architect escalation approval — extend iteration limit for a phase that has hit its cap. Records the approval decision in state and allows the blocked transition to proceed.
Use this skill when routing adversary findings in Phase 4. Provides routing logic, iteration limit enforcement, and escalation protocols.
Run Phase 4 (feedback routing) after an adversarial review FAIL. Routes adversary findings to the earliest affected phase, checks iteration limits, and escalates to human if limits are exceeded.
Use this skill during Phase 5 formal hardening. Provides tool selection, proof harness patterns, security/purity audit expectations, and verification result interpretation for Rust (Kani), Python (hypothesis), and TypeScript (fast-check).
Use this skill when committing VCSDD pipeline artifacts to git. Provides conventional commit message format, phase tag patterns, and atomic staging strategies.
Use this skill when evaluating VCSDD grading dimensions during adversarial review. Provides concrete evaluation checklists for each of the 5 dimensions and convergence signal detection.
Run Phase 5 (formal hardening) for the active VCSDD feature. Invokes vcsdd-verifier to execute proof obligations, security hardening, and a purity-boundary audit before convergence.
Run Phase 2b (implementation, Green phase) and Phase 2c (refactor) for the active VCSDD feature. Invokes vcsdd-builder to implement minimal code to pass failing tests, then refactor.
Initialize a VCSDD feature pipeline. Creates the .vcsdd/features/<name>/ directory tree, sets mode (strict|lean), and activates the feature. Must be run before any other VCSDD command.
Use this skill when applying VCSDD to Python projects. Provides hypothesis property testing patterns, mutmut mutation testing, and pytest integration for the VCSDD pipeline.
Use this skill when applying VCSDD to Rust projects. Provides Kani proof harness patterns, proptest strategies, cargo-fuzz integration, and cargo-mutants mutation testing guidance.
Use this skill when applying VCSDD to TypeScript/JavaScript projects. Provides fast-check property testing patterns, Stryker mutation testing, and vitest/jest integration for the VCSDD pipeline.
Use this skill when writing Phase 1a behavioral specifications in VCSDD. Provides EARS format requirements writing, purity boundary analysis, edge case enumeration, and spec quality validation patterns.
Run Phase 1c (spec review gate) for the active VCSDD feature. Spawns a fresh vcsdd-adversary instance to review the behavioral spec and verification architecture. Records PASS/FAIL gate verdict.
Run Phase 1a (behavioral specification) and Phase 1b (verification architecture) for the active VCSDD feature. Invokes vcsdd-builder to write EARS-format requirements and purity boundary analysis.
Use this skill when writing sprint contracts in VCSDD strict mode. Provides grading criteria patterns, dimension weight guidelines, and pass threshold formulation for the 5 review dimensions.
Display the current VCSDD pipeline status for the active (or specified) feature. Shows phase, mode, sprint count, open findings, pending proofs, and the traceability summary.
Run Phase 2a (test generation, Red phase) for the active VCSDD feature. Invokes vcsdd-builder to generate failing tests for all spec requirements. Records red phase evidence.
Display the full traceability chain for a VCSDD bead. Traverses the Chainlink graph from the given bead ID and shows all connected artifacts with status.
Use this skill when creating or querying VCSDD Chainlink bead traceability. Provides bead creation patterns, chain traversal, and completeness validation.
Use this skill when writing Phase 1b verification architecture documents. Provides purity boundary mapping, proof obligation design, and tier assignment guidance.
Full SDD (Spec-Driven Development) framework for Claude Code. 72+ slash commands (/dc:*), 14 reusable skills, 15 reasoning models, TDD as an iron law, and i18n support (ES/EN/PT). Turn requirements into tested, documented, production-ready code through 7 disciplined lifecycle phases.
Executes bash commands
Hook triggers when Bash tool is used
Modifies files
Hook triggers on file write and edit operations
Uses power tools
Uses Bash, Write, or Edit tools
Share bugs, ideas, or general feedback.
Spec-Driven Development Workflow for Claude Code. A 6-step pipeline: Requirements → Code Analysis → Design → Implement → Verify → Self-Improve. Specifications are the source of truth, code is a verified artifact.
Correctness workflow for structured development. 26 skills with configurable intensity levels: spec-before-code, skeptical review, enforced TDD, formal modeling, adversarial review, verification, documentation, and metrics. Intensity gates let you choose standard, high, or critical rigor per project.
Persona-driven AI development team: orchestrator, team agents, review agents, skills, slash commands, and advisory hooks for Claude Code
23 agent skills for systematic software development. Covers design, planning, TDD, code review, debugging, quality gates, and adversarial testing. Every skill is eval-tested with measured A/B deltas using Anthropic's skill evaluation framework.
SDLC enforcement for AI agents — TDD, planning, self-review, CI shepherd
Share bugs, ideas, or general feedback.