{"id":35591390,"url":"https://github.com/anton-abyzov/specweave","last_synced_at":"2026-06-02T04:00:45.466Z","repository":{"id":320981117,"uuid":"1083364012","full_name":"anton-abyzov/specweave","owner":"anton-abyzov","description":"Spec-first AI development: describe a feature → AI creates spec + plan + tasks, builds autonomously, syncs to GitHub/JIRA. Domain-expert skills for PM, Architect, Frontend, QA learn your patterns permanently. Claude Code, Codex, Cursor, Copilot \u0026 more.","archived":false,"fork":false,"pushed_at":"2026-05-13T06:57:22.000Z","size":198619,"stargazers_count":140,"open_issues_count":9,"forks_count":18,"subscribers_count":2,"default_branch":"develop","last_synced_at":"2026-05-13T08:35:58.888Z","etag":null,"topics":["ai","azure-devops","claude","cli","developer-tools","documentation","jira","testing","typescript"],"latest_commit_sha":null,"homepage":"https://spec-weave.com","language":"TypeScript","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/anton-abyzov.png","metadata":{"files":{"readme":"README.md","changelog":"CHANGELOG.md","contributing":".github/CONTRIBUTING.md","funding":null,"license":"LICENSE","code_of_conduct":"CODE_OF_CONDUCT.md","threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":"SECURITY.md","support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":"AGENTS.md","dco":null,"cla":null}},"created_at":"2025-10-25T21:31:34.000Z","updated_at":"2026-05-13T05:49:31.000Z","dependencies_parsed_at":"2026-04-02T11:07:59.314Z","dependency_job_id":null,"html_url":"https://github.com/anton-abyzov/specweave","commit_stats":null,"previous_names":["anton-abyzov/specweave"],"tags_count":688,"template":false,"template_full_name":null,"purl":"pkg:github/anton-abyzov/specweave","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/anton-abyzov%2Fspecweave","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/anton-abyzov%2Fspecweave/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/anton-abyzov%2Fspecweave/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/anton-abyzov%2Fspecweave/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/anton-abyzov","download_url":"https://codeload.github.com/anton-abyzov/specweave/tar.gz/refs/heads/develop","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/anton-abyzov%2Fspecweave/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":33805341,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-06-02T02:00:07.132Z","response_time":109,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"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":["ai","azure-devops","claude","cli","developer-tools","documentation","jira","testing","typescript"],"created_at":"2026-01-04T23:18:55.535Z","updated_at":"2026-06-02T04:00:45.445Z","avatar_url":"https://github.com/anton-abyzov.png","language":"TypeScript","funding_links":[],"categories":[],"sub_categories":[],"readme":"\u003ch1 align=\"center\"\u003eSpecWeave\u003c/h1\u003e\n\n\u003cp align=\"center\"\u003e\n  \u003cstrong\u003eAI-assisted development, under control.\u003c/strong\u003e\u003cbr/\u003e\n  Stop prompting. Start specifying.\n\u003c/p\u003e\n\n\u003cp align=\"center\"\u003e\n  \u003ca href=\"https://www.npmjs.com/package/specweave\"\u003e\u003cimg src=\"https://img.shields.io/npm/v/specweave?color=brightgreen\" alt=\"npm\" /\u003e\u003c/a\u003e\n  \u003cimg src=\"https://img.shields.io/badge/increments-600+-blue\" alt=\"600+ increments\" /\u003e\n  \u003cimg src=\"https://img.shields.io/badge/production_apps-10+-green\" alt=\"10+ production apps\" /\u003e\n  \u003cimg src=\"https://img.shields.io/badge/skills-100+-8B5CF6\" alt=\"100+ skills\" /\u003e\n  \u003cimg src=\"https://img.shields.io/badge/agent_platforms-49-orange\" alt=\"49 platforms\" /\u003e\n  \u003ca href=\"https://opensource.org/licenses/MIT\"\u003e\u003cimg src=\"https://img.shields.io/badge/License-MIT-yellow.svg\" alt=\"MIT\" /\u003e\u003c/a\u003e\n  \u003ca href=\"https://discord.gg/UYg4BGJ65V\"\u003e\u003cimg src=\"https://img.shields.io/badge/Discord-Join-5865F2?logo=discord\u0026logoColor=white\" alt=\"Discord\" /\u003e\u003c/a\u003e\n\u003c/p\u003e\n\n\u003cbr/\u003e\n\n## The Problem\n\n**36.82% of AI skills have security flaws** ([Snyk ToxicSkills](https://snyk.io/blog/toxicskills-malicious-ai-agent-skills-clawhub/)). In May 2025, 170 out of 1,645 vibe-coded apps had security vulnerabilities exposing personal data. No specs. No tests. No review. Just vibes.\n\nEvery alternative is an instruction layer — Cursor Rules, Copilot Instructions, Windsurf Rules, CLAUDE.md. They tell the AI *how* to write code but never *what* to build, never *how* to test it, and never *when* it's done.\n\nSpecWeave is a spec-first development layer. Configuration, not prompting. Enforced, not hoped for.\n\n\u003cbr/\u003e\n\n## The Solution\n\n```\nYou: \"Build a checkout flow with Stripe\"\n  ↓\n  spec.md → plan.md → tasks.md       ← you review the plan\n  ↓\n  Autonomous execution for hours      ← AI builds, tests, fixes\n  ↓\n  Quality gates (Grill + Judge-LLM)   ← code reviewed automatically\n  ↓\n  Synced to GitHub/JIRA/ADO           ← closed, documented, shipped\n```\n\nEvery feature starts as a specification — user stories, acceptance criteria, architecture decisions — before a single line of code is written. TDD enforces correctness. Quality gates catch what tests miss.\n\n\u003cbr/\u003e\n\n## Built With SpecWeave\n\n12 production projects shipped in 3 months. 5 in the App Store.\n\n| App | Platform | What It Does |\n|-----|----------|-------------|\n| [**EasyChamp**](https://easychamp.com) | Web (GCP) | Enterprise sports league management. 20+ microservices, ML video analytics. 4 years in production. |\n| [**SketchMate**](https://apps.apple.com/app/sketchmate-ai-draw-game/id6760250072) | App Store | AI drawing game — multi-model evaluation judges player art semantically. |\n| [**Lulla**](https://apps.apple.com/app/lulla-calm-baby-anywhere/id6756977992) | App Store | Baby sleep app with Apple Watch. ML cry classification (tired/hungry/pain). |\n| [**Football 2026**](https://apps.apple.com/app/football-2026-travel/id6757258711) | App Store + Web | World Cup 2026 companion. AI travel planner, live tickets, team stats. |\n| [**SkillUp Football**](https://apps.apple.com/app/skillup-football/id6756978002) | App Store | Coaches monetize training via Stripe. Instagram-like feed, scheduling. |\n| [**BizZone**](https://apps.apple.com/app/business-zone/id6756091030) | App Store | Student \u0026 business events with AI-powered news generation. |\n| [**EduFeed**](https://edufeed-jet.vercel.app/) | Web | NotebookLM meets Zoom. Upload videos, get quizzes, flashcards, live rooms. |\n| [**JobWeave**](https://jobweave.ai) | Web | AI-powered job search. Smart matching, resume optimization. |\n| [**SpecWeave**](https://github.com/anton-abyzov/specweave) | npm | The framework itself. 600+ increments, 538+ releases. |\n| [**SpecWeave Umbrella**](https://github.com/anton-abyzov/specweave-umb) | GitHub | Multi-repo orchestration workspace for all repositories. |\n| [**vskill**](https://github.com/anton-abyzov/vskill) | npm | Package manager for AI skills. Security scanning, 49 platforms. |\n| [**verified-skill.com**](https://verified-skill.com) | Web | Skill marketplace \u0026 studio. 105K+ verified skills, eval system. |\n\n[Browse increments on GitHub](https://github.com/anton-abyzov/specweave/tree/develop/.specweave/increments) — full transparency.\n\n\u003cbr/\u003e\n\n## Quick Start\n\n```bash\nnpm install -g specweave       # Node.js 20.12.0+\ncd your-project\nspecweave init .\n# Then in Claude Code, Cursor, Copilot, or any AI tool:\n# \"Build me a user authentication system\"\n```\n\n\u003cbr/\u003e\n\n## How It Compares\n\n| Capability | Cursor Rules | Copilot Instructions | Windsurf | Cline | Vibe Coding | **SpecWeave** |\n|------------|:---:|:---:|:---:|:---:|:---:|:---:|\n| Structured specs (spec + plan + tasks) | — | — | — | — | — | **Yes** |\n| Quality gates (Grill + Judge-LLM + 130 rules) | — | — | — | — | — | **Yes** |\n| Autonomous execution (hours, unattended) | — | — | — | — | — | **Yes** |\n| Multi-agent teams (parallel, contract-first) | — | — | — | — | — | **Yes** |\n| External sync (GitHub / JIRA / ADO) | — | — | — | — | — | **Yes** |\n| TDD enforcement (strict red-green-refactor) | — | — | — | — | — | **Yes** |\n| LSP code intelligence (198x faster) | — | — | — | — | — | **Yes** |\n| Self-improving skills (learns from corrections) | — | — | — | — | — | **Yes** |\n\nCursor tells AI \"use Tailwind.\" SpecWeave tells AI \"build a checkout flow with 5 acceptance criteria, test it, review it, sync to JIRA, and close.\"\n\n\u003cbr/\u003e\n\n## Key Features\n\n**Spec-First Planning** — Every feature starts as spec.md + plan.md + tasks.md. Configuration, not prompting.\n\n**TDD Enforcement** — Strict red-green-refactor. Tasks cannot close without passing tests. Coverage targets enforced.\n\n**Agent Swarms** — Run parallel agents across iTerm/tmux panes. Team lead splits work, each agent owns an increment.\n\n```\n┌──────────────────┬──────────────────┬──────────────────┐\n│  Agent 1 (auth)  │ Agent 2 (payments)│ Agent 3 (catalog)│\n│  sw:auto         │  sw:auto         │  sw:auto         │\n│  ████████░░ 80%  │  ██████░░░░ 60%  │  ████░░░░░░ 40%  │\n└──────────────────┴──────────────────┴──────────────────┘\n```\n\n**LSP Code Intelligence** — 198x faster than grep, 0 false positives. Semantic references, definitions, and types.\n\n**100+ Skills** — PM, Architect, QA, Security, DevOps, Frontend, Backend, Mobile, ML. Every skill is customizable via skill-memories without forking.\n\n**External Sync** — GitHub Issues, JIRA, Azure DevOps — bidirectional, real-time. Close an increment, external tools update automatically.\n\n**Enterprise Ready** — Compliance audit trails (SOC 2, HIPAA, FDA). Brownfield analysis. Multi-repo coordination. Multi-environment deployment.\n\n**Dashboard** — Built-in web dashboard for increment progress, analytics, cost tracking, and multi-project monitoring.\n\n\u003cbr/\u003e\n\n## Skills Ecosystem\n\nSpecWeave skills are published and verified at **[verified-skill.com](https://verified-skill.com)**. The [vskill](https://www.npmjs.com/package/vskill) package manager provides:\n\n- **Security scanning** — 52 attack patterns, SHA-256 pinning, blocklist API\n- **49 agent platforms** — one install deploys to Claude Code, Cursor, Copilot, Windsurf, and 45 more\n- **Skill evals** — unit tests, A/B comparisons, cross-model testing. Skills tested like programs.\n- **Visual Skill Studio** — `vskill eval serve` for benchmarks, comparisons, and history\n\n```bash\nnpx vskill install remotion-best-practices    # Install from registry\nnpx vskill eval run my-skill                  # Run eval suite\n```\n\n\u003cbr/\u003e\n\n## Core Commands\n\n| You say | SpecWeave runs |\n|---------|---------------|\n| \"Build me X\" | `sw:increment` → spec + plan + tasks |\n| \"Go ahead\" | `sw:auto` → autonomous execution |\n| \"Ship it\" | `sw:done` → quality gates + close |\n| \"Split into teams\" | `sw:team-lead` → parallel agents |\n| \"Review the code\" | `sw:code-reviewer` → 6 parallel reviewers |\n\n[Full command reference](https://spec-weave.com/docs/commands/overview)\n\n\u003cbr/\u003e\n\n## Upgrading to v1.1.0 (Opus 4.7 Alignment)\n\nSpecWeave v1.1.0 aligns the framework with Claude Opus 4.7 best practices.\n\n### P0 Fixes (required if using judge-llm or team-lead)\n\n1. **Remove fixed extended thinking**: `sw:judge-llm` no longer passes `thinking.budget_tokens`. Adaptive prompt hints replace it. No user action needed — the CLI handles this automatically.\n\n2. **Team-lead spawn heuristic**: `sw:team-lead` now only spawns agents when `domains ≥ 3` OR `tasks ≥ 15` OR `--parallel` is set. Previously it always spawned agents.\n\n3. **State markers removed**: `skill-chain-*.json` and `interview-*.json` state files are no longer written. `pretooluse-guard.ts` uses TaskGet-based state with filesystem fallback.\n\n### Config Migration\n\nRun the migration script to add new config keys with defaults:\n\n```bash\nspecweave migrate-config-0669\n```\n\n### Deprecations\n\nSee full list in [CHANGELOG.md](./CHANGELOG.md). Key migrations:\n- `/sw:tdd-red` → `/sw:tdd-cycle --phase red`\n- `/sw:github-sync` → `/sw-github:sync-spec`\n- `/sw:team-build` → `/sw:team-lead --preset \u003cname\u003e`\n\nFull migration guide: `.specweave/docs/internal/specs/opus-47-migration.md`\n\n## Documentation\n\n**[spec-weave.com](https://spec-weave.com)** — guides, reference, and enterprise docs.\n\n## Community\n\n[Discord](https://discord.gg/UYg4BGJ65V) · [YouTube](https://www.youtube.com/@antonabyzov) · [GitHub Issues](https://github.com/anton-abyzov/specweave/issues)\n\n## License\n\nMIT — [github.com/anton-abyzov/specweave](https://github.com/anton-abyzov/specweave)\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fanton-abyzov%2Fspecweave","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fanton-abyzov%2Fspecweave","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fanton-abyzov%2Fspecweave/lists"}