Specrew is a methodology for AI-assisted software delivery — a governance layer that runs on top of GitHub Spec Kit and Squad CLI and enforces the SDLC discipline that those tools alone don't enforce.
It is not a multi-agent framework, not an autopilot tool, and not another code generator. It is the layer that keeps the human in the loop at every decision boundary while letting agents do the work between boundaries.
Modern AI-assisted code tools optimize for throughput — finish more in less time. That works until the AI quietly decides things the human would have decided differently:
- Picks a database without asking
- Resolves an ambiguous requirement by guessing
- Skips a clarifying question to save a turn
- Crosses a planning-to-implementation boundary without authorization
- Ships work that "looks correct" but isn't traceable to a spec
Specrew was built after observing these failures empirically and concluding that the gap is not in the agent's capability; it is in the discipline around the agent. The same agent that auto-resolves a scope decision under one tool will surface it as a question under another. The difference is the methodology layer.
Specrew encodes that methodology as four guarantees:
- Boundary discipline. The lifecycle has explicit approval boundaries (
specify,clarify,plan,tasks,before-implement,review-signoff,retro,iteration-closeout,feature-closeout). One human authorization advances at most one boundary. No agent prose can simulate authorization. Enforcement is moving from prose to code (see Proposal 065, in flight as Feature 039). - Substantive interaction. Every boundary handoff is reviewable in the console with the essence of "what I just did / why I stopped / what I need from you" visible without opening files. Status pings are not enough.
- Audit-trail durability. Every verdict, decision, drift event, and bypass lives in
.squad/decisions.mdwith timestamps, commit hashes, and recognized verdict shapes. Sessions can be reconstructed after the fact; methodology lives in artifacts, not in agent memory. - Methodology survives the host. Specrew runs on GitHub Copilot CLI today. Claude Code, Codex CLI, and VS Code Chat are roadmap items (Proposal 069). The skill-level enforcement gates are host-agnostic by design — switching hosts must not weaken the methodology.
| If you want… | …use this instead |
|---|---|
| A multi-agent code library (orchestrate agents in Python) | CrewAI, AutoGen, LangGraph |
| Autopilot coding (let the agent run; check the output) | Devin, OpenInterpreter, Aider |
The spec-driven command surface alone (/speckit.specify, /speckit.plan, …) |
Spec Kit directly |
| The multi-agent runtime alone (specialist teams, agent charters) | Squad CLI directly |
| A code generator | None of these — Specrew is governance over agent-driven work, not a code generator |
Specrew composes Spec Kit + Squad into a methodology layer with enforced discipline. It is the smallest layer that keeps the human in control when agents are doing the typing.
Vanilla Spec Kit ships the slash-command surface but has no orchestration or boundary enforcement. Vanilla Squad runs multi-agent teams but doesn't drive a spec-driven lifecycle. Autopilot tools and multi-agent libraries optimize for throughput by letting the agent decide. Specrew goes the other direction: the spec is authoritative, drift is a first-class event, every boundary requires explicit human authorization, and the audit trail is durable. Different design point. Same agents.
- Active development line: 0.25.0
- Latest stable baseline: 0.24.3 (process-optimization bundle: closeout sync commands, markdown lint pre-boundary, validator memoization/parallelization/closed-iteration-index, repetition detector, PR-review integration)
- Alpha software, validated through dogfooding in this repository
- Built for a single developer today. Multi-developer reconciliation is a roadmap item (Proposal 010).
- Release truth lives in CHANGELOG.md, docs/versioning.md, and the
v0.NN.0tags.
specrew initbootstraps Spec Kit, Squad, and Specrew governance into a fresh or existing repospecrew startlaunches the canonical lifecycle session with handoff artifacts refreshedspecrew whererenders the velocity dashboard from canonical artifacts- The full lifecycle:
specify → clarify → plan → tasks → implement → review-signoff → retro → iteration-closeout → feature-closeout— with gate-respecting boundary stops by default (Proposal 066, shipped) - Session-state durability across reboots, worktree switches, and boundary events
- Slash-command catalog deployed to
.claude/skills/,.github/skills/, and.agents/skills/(Feature 024) - Validator memoization, parallelization, closed-iteration index, repetition detector — the v0.24.3 process-optimization bundle keeps the discipline cheap to enforce
- Reviewer-regression routing, session-loaded file change detection, drift-log integrity
- Pre-boundary markdown-lint auto-fix gate prevents lint round-trips at every boundary commit
- PR-review-integration soft warning surfaces missing
pr-review-resolution.mdwhen host has automated review available
- F-039 Launch-Mode Boundary Enforcement — mechanical refusal of agent boundary chaining (in flight, parked at iteration-closeout)
- F-040 Substantive Intake Questioning — persona-driven adaptive intake (next after F-039)
- Friction Dial (Proposal 100) — strict/default/autonomous modes for expert developers; composes Proposals 015 + 047 + 066
- Installed-File SDLC Audit (Proposal 099) — close the dogfooding deficit between maintainer paste-prompts and installed methodology files
- Multi-host launch (Proposal 069) — Claude Code and Codex CLI as alternatives to Copilot CLI
- Cost-aware model routing (Proposal 068) + Token Economy MVP — Junior tasks to cheap models, Senior tasks to strong models
See proposals/INDEX.md for the full proposal catalog (Shipped / Draft / Candidate).
Five minutes from zero to a running lifecycle session:
Install-Module Specrew -Scope CurrentUser -SkipPublisherCheck
mkdir C:\Dev\calculator && cd C:\Dev\calculator && git init
specrew init
specrew start "Build a web based calculator with only the + - * / MR MC M+ M- operations"See docs/getting-started.md for the full quickstart, install variants, and known limitations. See docs/user-guide.md for day-to-day usage.
| Platform | Status |
|---|---|
| Windows 11 (primary) | ✅ Fully validated |
| WSL Ubuntu | ✅ Manually validated end-to-end |
| Linux native (Ubuntu) | ✅ Path handling cross-platform; CI matrix configured |
| macOS | 🔧 Path handling cross-platform; CI matrix configured; no in-house validation yet |
- docs/getting-started.md — bootstrap + minimal flow
- docs/user-guide.md — day-to-day lifecycle usage
- docs/dashboard-guide.md — dashboard sections, flags, closeout snapshots
- docs/versioning.md — release-numbering policy and tag/changelog rules
- CHANGELOG.md — retroactive feature-release history
- proposals/INDEX.md — full proposal catalog (candidates, drafts, shipped)
- docs/roadmap-maintenance.md —
.specrew/roadmap.ymlmaintenance
Specrew is alpha. Reading, issues, and discussion are welcome now. External pull requests are intentionally deferred until the operating model and review boundaries stabilize. The dogfooding loop on this repository is the validation surface for every methodology change.
Specrew is released under the MIT License. See LICENSE for the repository license and NOTICE.md for upstream attribution covering derived Squad and Spec Kit materials.