Capability Reference

The Four Stations

Plan Forge is an AI-Native SDLC Forge Shop. Every capability on this page lives in one of four stations. See the full Shop Tour for deep-dive walkthroughs.

MCP Tools

These are the commands your AI assistant can run for you — in GitHub Copilot Chat, Claude, Cursor, or any MCP client. You don't call them by hand; you describe what you want in plain English and the assistant picks the right tool. New to Plan Forge? Just ask forge_capabilities first — it returns the whole map (tools, workflows, config, and a glossary) so the assistant knows everything that's available.

🧭 Forge-Master Studio (2 tools + dashboard — v2.63+, Phase-43 audit) open-ended reasoning · read-only · CTO defaults

A read-only reasoning orchestrator. Classifies user intent into one of 8 lanes (build, operational, troubleshoot, advisory, offtopic, tempering, principle-judgment, meta-bug-triage), retrieves OpenBrain memory context, and orchestrates other forge tools on the agent's behalf. Phase-43 flipped the CTO defaults on (observer enabled, L3 memory enabled, autoEscalate on, quorumAdvisory=auto), shipped lane-specific system-prompt overlays (advisory-cto, build-interviewer, troubleshoot-sre), and added the closed-loop forge_master_audit tool plus the pforge audit CLI. Phase-29 added the Forge-Master Studio dashboard tab with a curated prompt gallery, streaming chat, and a live tool-call trace pane. 📖 Manual: Forge-Master · 🗺️ Diagram: 3-stage intent classifier.

forge_master_ask

Accepts a freeform message. Returns a structured reasoning response built from intent classification, lane-specific prompt overlay, memory retrieval, and allowlist-gated read-only tool calls (≈38 read tools restored in Phase-43).

forge_master_audit (Phase-43)

Holistic CTO audit. Pulls drift, cost, open bugs, watcher alerts, deploy journal, and Crucible smelts; returns bounded report: summary, top 3 risks with evidence, P0/P1/P2 actions (≤5), cost note. Drives the weekly digest and incident-triggered review.

Studio tab · prompt gallery · chat stream · tool-call trace

Dashboard UI at localhost:3100/dashboard. Also available as CLI via pforge forge-master status|logs and the new pforge audit [--since --tier --schedule --on-incident].

Autonomous Execution

📖 Manual: Advanced Execution · 🗺️ Diagram: parallel slice DAG

Agent-Per-Slice Routing

Assign a different AI model to each execution role. The orchestrator auto-selects based on the current operation — tune cost vs. quality at every stage without changing your plan files.

📖 Manual: Multi-Agent Routing · 🗺️ Diagram: host-aware routing

Auto-Escalation

When a slice fails on one model, the orchestrator automatically walks the escalationChain and retries on the next model — no manual intervention. Emits a slice-escalated event on each re-route.

📖 Manual: Advanced Execution · 🗺️ Diagram: escalation chain

Model Performance Tracking

Per-slice performance data is appended to .forge/model-performance.json after every run. The orchestrator reads this on startup and auto-selects the cheapest model with >80% historical success rate for each slice type.

📖 Manual: Cost & Economics

Quorum Mode

Multi-model consensus: dispatch complex slices to 3 AI models for independent analysis, synthesize the best approach, then execute with higher confidence. A/B tested: +20% more tests, better code structure, fewer brittle patterns vs single-model execution. Read the full A/B test results →

📖 Manual: Quorum Mode · 🗺️ Diagram: quorum flow · complexity rubric

Web UI — Live Dashboard

localhost:3100/dashboard — 38 real-time tabs via WebSocket, grouped into Forge, LiveGuard, Forge-Master, and Settings. No build step. Also runs standalone: node pforge-mcp/server.mjs --dashboard-only (8 core tabs shown below)

📖 Manual: The Dashboard · 🗺️ Diagram: tab taxonomy (38 tabs)

Agents & Skills

📖 Manual: Instructions & Agents · Multi-Agent Mode

Temper Guards & Warning Signs — Every instruction file includes tables of common shortcuts agents use (with rebuttals) and observable anti-patterns that indicate the file's guidance is being violated.

Observability & Memory

📖 Manual: Memory Architecture · 🗺️ Diagram: three-tier memory capture

Stack Presets

📖 Manual: Customization & Presets

REST API — External Integration

The MCP server exposes a REST API for external agents, CI systems, and tools like OpenClaw. Discover the full surface via GET /api/capabilities or GET /.well-known/plan-forge.json on first connect.

📖 Manual: REST API Reference · 🗺️ Diagram: integration surfaces

Full curl examples and config template: AGENT-SETUP.md Section 6.

Bridge — External Notifications

The Plan Forge Bridge subscribes to the WebSocket hub and dispatches run events to external platforms. Rate-limited (1/5s per channel), with automatic reconnect.

📖 Manual: Remote Bridge · 🗺️ Diagram: bridge fan-out

{
  "bridge": {
    "enabled": true,
    "channels": [
      { "type": "telegram", "url": "https://api.telegram.org/bot<TOKEN>/sendMessage", "chatId": "<ID>", "level": "important" },
      { "type": "slack",    "url": "https://hooks.slack.com/services/...", "level": "all" },
      { "type": "discord",  "url": "https://discord.com/api/webhooks/...", "level": "critical" },
      { "type": "webhook",  "url": "https://your-endpoint.example.com/hook", "level": "all" }
    ]
  }
}

CI/CD Hook Event

The ci-triggered event is emitted when a CI workflow is dispatched from a plan run. Observable via the WebSocket hub or captured in the run's events.log. The slice-escalated event is emitted when a slice is re-routed to a new model via the escalation chain.

📖 Manual: Event Catalog

Updating an Existing Install

pforge smith automatically checks GitHub for a newer Plan Forge release — 5 s timeout, 24 h cache in .forge/version-check.json, silent when offline.

pforge self-update --force  # latest GitHub release
pforge update              # auto-mode (v2.56.0+)
pforge update --from-github # force GitHub tag

git clone https://github.com/srnichols/plan-forge.git

Control the update source with pforge config set update-source <auto|github-tags|local-sibling> (v2.56.0+). See Manual Appendix G.

Dual-Publish Extensions

pforge ext publish <path> validates the extension and outputs two catalog entries in one command.

GitHub Stack Integration

First-class integration with GitHub Copilot, GitHub Models, and GitHub Actions for cloud-based execution and security-driven plan generation.

📖 Manual: Plan Forge on the GitHub Stack · 🗺️ Diagram: GitHub stack architecture