Skip to content

positioning.md

Latest commit

 

History

History
194 lines (155 loc) · 8.85 KB

positioning.md

File metadata and controls

194 lines (155 loc) · 8.85 KB

Nexo Spec Kit — Product Positioning

One-liner: The structured AI development toolkit that turns ideas into executable specifications — with Phase 0 ideation, enterprise controls, and programmatic governance built in.

Problem Statement

Modern software teams using AI coding assistants face a structural gap: AI agents are excellent at implementing but poor at deciding what to build. Without a structured front-end discovery phase, teams accumulate poorly-scoped features, duplicated context, and brittle handoffs between AI agents and human reviewers.

Existing specification toolkits focus on the how (spec templates, plan documents) but skip the why (ideation, strategic selection, scope validation). This forces teams to bolt on ad-hoc ideation steps or skip them entirely.

Target Users

Segment Description Primary Pain
AI-native engineering teams Teams using 2+ AI coding agents simultaneously Agent context fragmentation, duplicated effort
Product + Engineering hybrids Teams where PMs co-author specifications Tooling that handles only dev-facing workflows
Enterprise software shops Organizations with compliance, provenance, and audit requirements No structured record-keeping for AI decisions
Open-source maintainers Projects adopting AI-assisted contribution workflows Missing governance layer for AI PRs

Elevator Pitch

Nexo Spec Kit is an independent toolkit and CLI for teams who build software with AI agents. It extends Spec-Driven Development with a Phase 0 discovery layer that runs before any spec is written:

  1. Phase 0 — Ideate: Structured AI ideation using SCAMPER + HMW to generate an ideas backlog. Select a winner with AI-RICE scoring. Complete the Integrated AI Vision Canvas and vision brief. Gate G0 validation.
  2. SDD Core — Specify: Author executable specifications that constrain agents to the right scope (/speckit.constitution, /speckit.specify, /speckit.clarify).
  3. SDD Core — Plan and implement: Generate traceable implementation plans and delegate execution to AI agents (/speckit.plan, /speckit.tasks, /speckit.implement, /speckit.amend).
  4. Enterprise layer: Scope gate contracts, handoff metadata, orchestration contracts, and a compliance checker create an auditable trail from idea to shipped code.

Nexo Spec Kit integrates with 18 AI coding agents (Claude Code, Gemini CLI, GitHub Copilot, Cursor, Qwen Code, opencode, Codex CLI, Windsurf, Kilo Code, Auggie CLI, Roo Code, CodeBuddy, Qoder CLI, Amazon Q Developer CLI, Amp, SHAI, Antigravity, IBM Bob) and remains upstream-compatible with github/spec-kit.

Differentiation Matrix

The following table compares Nexo Spec Kit with the upstream github/spec-kit project based on capabilities verifiable in each repository's source code and documentation at the time of writing.

Capability Upstream github/spec-kit Nexo Spec Kit
Phase 0 AI ideation (IDEATE → SELECT → STRUCTURE → VALIDATE) ❌ ✅ Full workflow with state checks and G0 gate
/speckit.amend command ❌ ✅ Post-implementation spec amendment
specify update (CLI update check) ❌ ✅
specify hierarchy-contract ❌ ✅ Program/Epic/Feature hierarchy validation
specify scope-detect ❌ ✅ Adaptive scope detection with gate-ready JSON
specify scope-gate ❌ ✅ Decomposition gate with follow/inspect/override
specify productivity cockpit ❌ ✅ start and update task management workflows
Scope gate contract module ❌ ✅ Typed strict/non-strict validation
Handoff metadata schema ❌ ✅ Structured handoff with provenance tracking
Orchestration contract module ❌ ✅ Typed orchestration contract validation
Enterprise compliance checker ❌ ✅ License, legal markers, trademark policy
Release deploy pipeline ❌ ✅ Automated VM deploy + Google Drive backup
Upstream compatibility opt-in N/A ✅ --template-repo github/spec-kit
Agent integrations ✅ 18 named + "generic" catch-all ✅ 18 named agents
Extension system ✅ ✅
MIT license ✅ ✅ Preserved with attribution

Value Pillars

1. Structured Discovery (Phase 0)

AI agents work best when given clear, bounded objectives. Phase 0 transforms vague ideas into ranked, scoped feature candidates before a single spec is written. Commands: /speckit.ideate, /speckit.select, /speckit.structure, /speckit.validate.

Key artifacts:

  • .spec-kit/ideas_backlog.md
  • .spec-kit/idea_selection.md
  • .spec-kit/ai_vision_canvas.md
  • .spec-kit/vision_brief.md
  • .spec-kit/approvals/g0-validation-report.md

2. Executable Specifications

Specifications are not documents — they are contracts between human intent and AI execution. The SDD workflow produces .specify/specs/<feature>/spec.md, plan.md, and tasks.md as artifacts that agents directly consume.

Commands: /speckit.constitution, /speckit.specify, /speckit.clarify, /speckit.plan, /speckit.tasks, /speckit.implement, /speckit.amend, /speckit.analyze, /speckit.checklist, /speckit.taskstoissues.

3. Enterprise-Grade Governance

Every AI decision is traceable. The governance layer includes:

  • Scope gate contracts (specify scope-gate, specify scope-detect): programmatic follow/inspect/override decisions with audit JSON.
  • Handoff metadata schema: structured handoff records with provenance.
  • Orchestration contract module: strict/non-strict typed validation.
  • Hierarchy contract (specify hierarchy-contract): normalizes Program/Epic/Feature hierarchy payloads.
  • Compliance checker: verifies license, legal markers, and trademark policy before each release.
  • Provenance playbook: upstream intake with git cherry-pick -x for full commit-level traceability.

4. Broad Agent Integration Surface

18 AI coding agents are supported out of the box:

Key Agent Type
claude Claude Code CLI
gemini Gemini CLI CLI
copilot GitHub Copilot IDE
cursor-agent Cursor IDE
qwen Qwen Code CLI
opencode opencode CLI
codex Codex CLI CLI
windsurf Windsurf IDE
kilocode Kilo Code IDE
auggie Auggie CLI CLI
roo Roo Code IDE
codebuddy CodeBuddy CLI
qoder Qoder CLI CLI
q Amazon Q Developer CLI CLI
amp Amp CLI
shai SHAI CLI
agy Antigravity IDE
bob IBM Bob IDE

New agents are added via a documented, convention-driven integration pattern (see AGENTS.md). Use specify init --ai <key> to bootstrap any agent.

Promise and Non-Goals

Promise

  • Every project bootstrapped with specify init has a documented Phase 0 artifact and a spec-ready directory structure within minutes.
  • Every AI agent interaction is scoped, traceable, and reproducible.
  • Enterprise teams can adopt Nexo Spec Kit without sacrificing compliance or audit requirements.

Non-Goals

  • Not a code generator: Nexo Spec Kit orchestrates AI agents; it does not generate application code itself.
  • Not a CI/CD system: Release and deploy workflows are lightweight helpers, not a replacement for a full platform engineering stack.
  • Not a managed SaaS: This is an open-source CLI and template toolkit. No hosted infrastructure is required or provided.
  • Not a GitHub product: This toolkit is not affiliated with, endorsed by, or supported by GitHub. See docs/trademarks.md.

Compatibility Statement

Nexo Spec Kit is upstream-compatible with github/spec-kit:

  • specify init pulls templates from nsalvacao/spec-kit by default.
  • Set SPECIFY_TEMPLATE_REPO=github/spec-kit or pass --template-repo github/spec-kit to use upstream templates.
  • All enhancements are additive and isolated. No upstream behavior is removed or replaced.
  • Upstream intake is governed by the 7-step workflow in AGENTS.md, with provenance tracking via git cherry-pick -x.

For the compatibility policy detail, see docs/compatibility.md. For upstream sync procedures, see docs/upstream-sync.md.

Related Documents