{"id":45166011,"url":"https://github.com/ikennaokpala/forge","last_synced_at":"2026-02-20T07:02:08.652Z","repository":{"id":339511445,"uuid":"1152006112","full_name":"ikennaokpala/forge","owner":"ikennaokpala","description":"Forge is an autonomous behavioural validation engineering swarm that treats quality as something forged into software continuously, not bolted on at the end.","archived":false,"fork":false,"pushed_at":"2026-02-20T06:47:12.000Z","size":149,"stargazers_count":7,"open_issues_count":14,"forks_count":2,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-02-20T06:57:34.989Z","etag":null,"topics":["agent-skill","bdd","claude-code","gherkin","quality-engineering","testing"],"latest_commit_sha":null,"homepage":"","language":null,"has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/ikennaokpala.png","metadata":{"files":{"readme":"README.md","changelog":"CHANGELOG.md","contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2026-02-07T07:58:30.000Z","updated_at":"2026-02-20T06:45:06.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/ikennaokpala/forge","commit_stats":null,"previous_names":["ikennaokpala/forge"],"tags_count":3,"template":false,"template_full_name":null,"purl":"pkg:github/ikennaokpala/forge","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ikennaokpala%2Fforge","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ikennaokpala%2Fforge/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ikennaokpala%2Fforge/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ikennaokpala%2Fforge/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/ikennaokpala","download_url":"https://codeload.github.com/ikennaokpala/forge/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ikennaokpala%2Fforge/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":29643960,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-02-20T05:21:04.652Z","status":"ssl_error","status_checked_at":"2026-02-20T05:21:04.238Z","response_time":59,"last_error":"SSL_read: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["agent-skill","bdd","claude-code","gherkin","quality-engineering","testing"],"created_at":"2026-02-20T07:02:08.070Z","updated_at":"2026-02-20T07:02:08.643Z","avatar_url":"https://github.com/ikennaokpala.png","language":null,"funding_links":[],"categories":[],"sub_categories":[],"readme":"# Forge\n\n**Behavioral validation forged in, not bolted on.**\n\nForge is an autonomous behavioral validation swarm skill for [Claude Code](https://claude.com/claude-code) that combines BDD behavioral verification, 7 behavioral validation gates, confidence-tiered learning, topological governance, and self-healing fix loops. It spawns 8 specialized agents that work in parallel to verify, test, fix, and commit — continuously — until every Gherkin scenario passes and every behavioral validation gate clears.\n\n---\n\n## Key Features\n\n- **8 specialized agents** working in parallel with cost-optimized model routing\n- **Gherkin behavioral specifications** as the single source of truth\n- **7 behavioral validation gates**: Functional, Behavioral, Coverage, Security, Accessibility, Resilience, Contract\n- **12 topological governance specifications** (§1.1–§1.12) — mathematical foundations for autonomous behavioral validation\n- **Confidence-tiered fix patterns** (Platinum/Gold/Silver/Bronze) with Nash Equilibrium convergence\n- **Defect prediction** based on historical failure data and file changes\n- **Chaos/resilience testing** with controlled failure injection\n- **Cross-context dependency awareness** with cascade re-testing and sheaf cohomology consistency\n- **Shared types and cross-cutting validation** across bounded contexts\n- **Agent-optimized ADRs** with MUST/MUST NOT constraints and verification commands\n- **Visual regression testing** with pixel-by-pixel comparison\n- **Architecture-agnostic** — monolith, microservices, monorepo, mobile+backend\n- **Optional Agentic QE integration** for enhanced pattern search, security scanning, and more\n- **External-only mocking** — mock third-party services, never internal code (production-validated policy)\n- **Spec drift detection** — detects when Gherkin specs and implementation diverge\n- **LLM-as-Judge meta-review** — second-model evaluation with Anti-Echo-Chamber guarantee\n- **Self-reflection gate** — Bug Fixer asks \"What could go wrong?\" before committing\n- **Hallucination Gate** — deterministic pre-LLM boundary (AST resolution, contract hash, mocking detection)\n- **Agent criticality scoring** — bottleneck detection via Dirichlet energy and automatic optimization\n- **Narya-proofs** — counterfactual verification proving fix necessity and sufficiency\n- **Property-based testing** — generate 1000+ test cases from invariants\n- **Mutation testing** — inject bugs to verify test effectiveness\n- **Blake3 witness chain** — cryptographic tamper-evident audit trail for gate verdicts\n- **Infrastructure readiness markers** — specify formally, implement pragmatically, upgrade transparently\n\n---\n\n## Philosophy\n\n### Three Pillars\n\n| Pillar | Source | What It Does |\n|--------|--------|--------------|\n| **Build** | DDD+ADR+TDD methodology | Structured development with behavioral validation gates, defect prediction, confidence-tiered fixes |\n| **Verify** | BDD/Gherkin behavioral specs | Continuous behavioral verification — the PRODUCT works, not just the CODE |\n| **Heal** | Autonomous E2E fix loop | Test → Analyze → Fix → Commit → Learn → Repeat |\n\n### \"DONE DONE\"\n\n\"DONE DONE\" means: the code compiles AND the product behaves as specified. Every Gherkin scenario passes. Every behavioral validation gate clears. Every dependency graph is satisfied.\n\n---\n\n## Quick Start\n\n```bash\n# Copy SKILL.md to your Claude Code skills directory\ncp SKILL.md ~/.claude/skills/forge.md\n\n# Run on your project\n/forge --autonomous --context payments\n```\n\n---\n\n## Invocation Modes\n\n| Command | Description |\n|---------|-------------|\n| `/forge --autonomous --all` | Full autonomous run — all contexts, all gates |\n| `/forge --autonomous --context [name]` | Single context autonomous run |\n| `/forge --verify-only` | Behavioral verification only (no fixes) |\n| `/forge --verify-only --context [name]` | Verify single context |\n| `/forge --fix-only --context [name]` | Fix failures, don't generate new tests |\n| `/forge --learn` | Analyze patterns, update confidence tiers |\n| `/forge --add-coverage --screens [names]` | Add coverage for new screens/pages/components |\n| `/forge --spec-gen --context [name]` | Generate Gherkin specs for a context |\n| `/forge --spec-gen --all` | Generate Gherkin specs for all contexts |\n| `/forge --gates-only` | Run behavioral validation gates without test execution |\n| `/forge --gates-only --context [name]` | Run behavioral validation gates for single context |\n| `/forge --predict` | Defect prediction only |\n| `/forge --predict --context [name]` | Predict defects for single context |\n| `/forge --chaos --context [name]` | Chaos/resilience testing for a context |\n| `/forge --chaos --all` | Chaos testing for all contexts |\n| `/forge --drift-check` | Spec drift detection |\n| `/forge --drift-check --context [name]` | Drift check for single context |\n| `/forge --regressions` | Behavioral regression analysis |\n| `/forge --regressions --context [name]` | Regressions for single context |\n| `/forge --meta-review` | LLM-as-Judge meta-evaluation |\n| `/forge --meta-review --context [name]` | Meta-review for single context |\n| `/forge --mutation --context [name]` | Mutation testing for a context |\n| `/forge --mutation --critical-only` | Mutation testing for critical paths only |\n\n---\n\n## Architecture\n\n### Autonomous Loop\n\n```\nSpecify → Test → Analyze → Fix → Audit → Gate → Commit → Learn → Repeat\n```\n\n```\n┌────────────────────────────────────────────────────────────────────┐\n│                    FORGE AUTONOMOUS LOOP                            │\n├────────────────────────────────────────────────────────────────────┤\n│                                                                    │\n│  ┌──────────┐   ┌──────────┐   ┌──────────┐   ┌──────────┐      │\n│  │ Specify  │──▶│   Test   │──▶│ Analyze  │──▶│   Fix    │      │\n│  │ (Gherkin)│   │ (Run)    │   │ (Root    │   │ (Tiered) │      │\n│  └──────────┘   └──────────┘   │  Cause)  │   └──────────┘      │\n│       ▲                        └──────────┘        │              │\n│       │                                            ▼              │\n│  ┌──────────┐   ┌──────────┐   ┌──────────┐   ┌──────────┐      │\n│  │  Learn   │◀──│  Commit  │◀──│  Gate    │◀──│  Audit   │      │\n│  │ (Update  │   │ (Auto)   │   │ (7 Gates)│   │ (A11y)   │      │\n│  │  Tiers)  │   └──────────┘   └──────────┘   └──────────┘      │\n│  └──────────┘                                                     │\n│       │                                                           │\n│       └──────────────── REPEAT ──────────────────────────────────│\n│                                                                    │\n│  Loop continues until: ALL 7 VALIDATION GATES PASS or MAX 10    │\n└────────────────────────────────────────────────────────────────────┘\n```\n\n### Execution Phases\n\n1. **Phase 0** — Backend setup (build, run, health check, seed data)\n2. **Phase 1** — Behavioral specification \u0026 architecture records (Gherkin specs, ADRs)\n3. **Phase 2** — Contract \u0026 dependency validation (schemas, shared types, cross-cutting)\n4. **Phase 3** — Swarm initialization (load patterns, predictions, confidence tiers)\n5. **Phase 4** — Spawn 8 autonomous agents in parallel\n6. **Phase 5** — Behavioral validation gates evaluation (7 gates after every fix cycle, BFT consensus ≥5/7)\n\n---\n\n## Behavioral Validation Gates\n\n| Gate | Check | Threshold | Blocking |\n|------|-------|-----------|----------|\n| 1. Functional | All tests pass | 100% pass rate | YES |\n| 2. Behavioral | Gherkin scenarios satisfied | 100% of targeted scenarios | YES |\n| 3. Coverage | Path coverage | \u003e=85% overall, \u003e=95% critical | YES (critical only) |\n| 4. Security | No secrets, SAST checks, no injection vectors | 0 critical/high violations | YES |\n| 5. Accessibility | Labels, target sizes, contrast | WCAG AA | Warning only |\n| 6. Resilience | Offline, timeout, error handling | Tested for target context | Warning only |\n| 7. Contract | API response matches schema | 0 mismatches | YES |\n\n---\n\n## Agent Roles\n\n| Agent | Model | Role | v1.2.0 Enhancement |\n|-------|-------|------|--------------------|\n| **Specification Verifier** | Sonnet | Generates/validates Gherkin specs and ADRs for bounded contexts | — |\n| **Test Runner** | Haiku | Executes E2E test suites, parses results, maps failures to specs | — |\n| **Failure Analyzer** | Sonnet | Root cause analysis, pattern matching, dependency impact assessment | MaTTS — 3 parallel reasoning trajectories with self-contrast |\n| **Bug Fixer** | Opus | Applies confidence-tiered fixes from first principles | Driver-Observer algebraic connectivity (λ₂ monitoring) |\n| **Behavioral Validation Gate Enforcer** | Haiku | Evaluates all 7 gates, arbitrates agent disagreements | BFT consensus model (≥5/7 threshold, VETO for blocking gates) |\n| **Accessibility Auditor** | Sonnet | WCAG AA audit: labels, contrast, targets, focus order | — |\n| **Auto-Committer** | Haiku | Stages fixed files, creates detailed commits with gate statuses | — |\n| **Learning Optimizer** | Sonnet | Updates confidence tiers, defect prediction, coverage metrics | DISTILL phase — LoRA-style abstraction with EWC++ anti-forgetting |\n\n---\n\n## Topological Governance (v1.2.0)\n\nForge v1.2.0 introduces 12 formal topological governance specifications (§1.1–§1.12) that provide mathematical foundations for autonomous behavioral validation. Production heuristics from v1.1.0 — criticality scoring, regression tracking, blocking gates — are now anchored to formal mathematical equivalents.\n\n### Four Specification Clusters\n\n| Cluster | Sections | Purpose |\n|---------|----------|---------|\n| **Consistency \u0026 Verification** | §1.1–§1.5 | Sheaf cohomology for cross-context consistency, Dirichlet energy for system tension, persistent Laplacian for regression tracking, Hallucination Gate for pre-LLM verification, Blake3 witness chain for tamper-evident audit |\n| **Swarm Stability** | §1.6–§1.7 | Algebraic connectivity (Fiedler value λ₂) for agent coordination monitoring, MinCut isolation for quarantining anomalous agent output |\n| **Memory \u0026 Reasoning** | §1.8–§1.11 | Hyperbolic memory (Poincaré ball) for hierarchical code embeddings, GF(3) triadic validation for phase transitions, Narya-proofs for counterfactual fix verification, Johnson-Lindenstrauss for sublinear test coverage |\n| **Execution Plane** | §1.12 | WASM/Rust pure-function tasks for deterministic verification (Blake3 hashing, eigenvalue computation, GF(3) validation, HNSW search, contract hash comparison, JL projection) |\n\n### Infrastructure Readiness\n\nEvery specification is operational today. Infrastructure readiness markers define the path from \"correct\" to \"correct and fast\":\n\n| Specification | Current Implementation | Native Infrastructure |\n|---|---|---|\n| Blake3 witness chain (§1.5) | SHA-256 hashing | Blake3 native hashing |\n| Hyperbolic memory (§1.8) | Flat key-value lookups across 10 namespaces | HNSW-indexed Poincaré ball embeddings |\n| JL coverage (§1.11) | Defect prediction with failure probability ranking | Random projection to O(log n) representative tests |\n| WASM execution (§1.12) | LLM structured reasoning for pure functions | WASM/Rust compilation with sub-ms latency |\n\n---\n\n## Configuration\n\n### Project Config (optional)\n\n```yaml\n# forge.config.yaml — placed at repo root\narchitecture: microservices\nbackend:\n  services:\n    - name: auth-service\n      port: 8081\n      healthEndpoint: /health\n      buildCommand: npm run build\n      runCommand: npm start\nfrontend:\n  technology: react\n  testCommand: npx cypress run --spec {target}\n  testDir: cypress/e2e/\n  specDir: cypress/e2e/specs/\n\n# Model routing overrides\nmodel_routing:\n  bug-fixer: opus\n  failure-analyzer: sonnet\n  test-runner: haiku\n\n# Visual regression\nvisual_regression:\n  enabled: true\n  threshold: 0.001\n\n# Agentic QE integration\nintegrations:\n  agentic-qe:\n    enabled: true\n    domains: [defect-intelligence, security-compliance, visual-accessibility, contract-testing]\n```\n\n### Context Config (optional)\n\n```yaml\n# forge.contexts.yaml — bounded context definitions\ncontexts:\n  - name: identity\n    testFile: identity.cy.ts\n    specFile: identity.feature\n    paths: 68\n    subdomains: [Auth, Profiles, Verification]\n  - name: payments\n    testFile: payments.cy.ts\n    specFile: payments.feature\n    paths: 89\n    subdomains: [Wallet, Cards, Transactions]\n\ndependencies:\n  identity:\n    blocks: [payments, orders]\n  payments:\n    depends_on: [identity]\n    blocks: [orders, subscriptions]\n```\n\nIf no configuration files are present, Forge auto-discovers the project structure on first run.\n\n---\n\n## Agentic QE Integration\n\nForge optionally integrates with [Agentic QE](https://github.com/proffesor-for-testing/agentic-qe) via MCP for enhanced capabilities:\n\n| Capability | Without AQE | With AQE |\n|-----------|-------------|----------|\n| Pattern Storage | claude-flow memory | ReasoningBank (vector-indexed, 150x faster) |\n| Defect Prediction | File changes + history | Specialized defect-intelligence agents |\n| Security Scanning | Gate 4 static checks | Full SAST/DAST analysis |\n| Accessibility | Built-in auditor | visual-tester + accessibility-auditor |\n| Contract Testing | Schema validation | contract-validator + graphql-tester |\n| Progress | `.forge/progress.jsonl` | AG-UI real-time streaming |\n\nAll AQE features are additive. Forge works identically without AQE installed.\n\n---\n\n## References\n\n- [Continuous Behavioral Verification: Ongoing Path to Done](https://www.linkedin.com/pulse/continuous-behavioral-verification-ongoing-path-done-ikenna-okpala) — Ikenna Okpala\n- [Build with Quality Skill: How I Build Software 10x Faster](https://www.linkedin.com/pulse/build-quality-skill-how-i-build-software-10x-faster-mondweep-chakravorty) — Mondweep Chakravorty\n- [claude-code-v3-qe-skill](https://github.com/mondweep/vibe-cast) — V3 QE Skill\n- [agentic-qe](https://github.com/proffesor-for-testing/agentic-qe) — Agentic QE Framework\n- Advanced Topological Governance in Autonomous Software Engineering — Formal mathematical foundations (sheaf theory, spectral analysis, Galois fields) for v1.2.0 specifications\n\n---\n\n## License\n\nMIT\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fikennaokpala%2Fforge","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fikennaokpala%2Fforge","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fikennaokpala%2Fforge/lists"}