Spec Kit + Plan Forge — Better Together

Why Use Both?

🌱

Spec Kit

Defines what to build. Turns ideas into executable specifications via slash commands (/speckit.specify, /speckit.plan). 85K+ stars, 25+ AI agents, 40+ community extensions.

Strengths: Spec-first methodology, broad agent coverage, massive community

🔨

Plan Forge

Enforces how it's built. Hardens specs into locked execution contracts with 18 guardrail files, 19 reviewer agents, lifecycle hooks, and independent review sessions.

Strengths: Runtime enforcement, enterprise guardrails, scope-contract locking

The Combined Workflow

1. Define with Spec Kit

/speckit.constitution → /speckit.specify → /speckit.plan → /speckit.tasks

Output: specs/feature/spec.md, plan.md, tasks.md, memory/constitution.md

2. Plan Forge Auto-Detects

Step 0 (Specifier) scans the project and finds Spec Kit artifacts. Offers to import them directly — no re-specifying needed.

Maps: spec.md → feature spec, plan.md → execution contract, constitution.md → PROJECT-PRINCIPLES.md

3. Harden → Execute → Review → Ship

Plan Forge runs pre-flight checks (Step 1), locks the scope into an execution contract (Step 2), builds slice by slice with guardrails (Step 3), sweeps for completeness (Step 4), runs independent review (Step 5), and ships with confidence (Step 6).

The spec becomes an enforceable contract — not a suggestion the AI can ignore.

Shared Extension Model

Plan Forge's extension catalog uses the same format as Spec Kit's catalog.community.json. Extensions marked speckit_compatible: true work in both tools.

Spec Kit

specify extension search
specify extension add <name>

Plan Forge

pforge ext search
pforge ext add <name>

Same commands, same catalog model, shared extensions. Build once, use in both.

MCP bonus: Plan Forge's MCP server exposes 8 forge tools (forge_smith, forge_sweep, forge_diff, etc.) as native function calls. Composable with OpenBrain's MCP server in the same config — your agent gets forge operations + persistent memory in one session.

Artifact Mapping

Spec Kit Artifact	Plan Forge Equivalent	Auto-Import?
`specs/feature/spec.md`	Phase Plan — specification section	✅ Step 0
`specs/feature/plan.md`	Hardened execution contract	✅ Step 0
`specs/feature/tasks.md`	Execution slices	⚡ Step 2
`memory/constitution.md`	`docs/plans/PROJECT-PRINCIPLES.md`	✅ Path D
`/speckit.clarify` output	NEEDS CLARIFICATION markers resolved	⚡ Step 0
`/speckit.analyze` report	Step 5 Review Gate input	Manual

When to Use Which

Pick Spec Kit if...

• You use multiple AI tools (not just VS Code)
• You want the largest community and extension ecosystem
• You prefer a lightweight spec-first methodology
• GitHub's backing matters for long-term viability

Pick Plan Forge if...

• You want deep guardrails that auto-enforce during coding
• You need specialized reviewer agents (18 built-in)
• You care about scope-contract locking and drift prevention
• You're building enterprise/production software

Or use both

Spec Kit writes the spec. Plan Forge enforces it. Extensions work in both. You get the best of each world.

Quick Start: Combined Workflow

# 1. Spec Kit: define what to build
/speckit.constitution Create principles focused on code quality and testing
/speckit.specify Build a task management API with auth and real-time updates
/speckit.plan Use .NET 9, PostgreSQL, SignalR
/speckit.tasks

# 2. Plan Forge: set up guardrails (auto-detects Spec Kit artifacts)
.\setup.ps1 -Preset dotnet -Agent claude

# 3. Plan Forge: Specifier imports Spec Kit spec automatically
#    → "I found Spec Kit artifacts. Import spec.md as execution contract?"

# 4. Plan Forge: harden, execute, review, ship
#    Scope locked. Guardrails loaded. Independent review. Ship with confidence.