Commands Reference
Core Commands (forgeplan-workflow)
Section titled “Core Commands (forgeplan-workflow)”/forge “task description”
Section titled “/forge “task description””Plugin: forgeplan-workflow
Runs the full Forgeplan methodology cycle: Route → Shape → Validate → Build → Evidence → Activate.
/forge "Add OAuth2 authentication"The command determines depth automatically and creates the right artifacts.
/forge-cycle PRD-XXX
Section titled “/forge-cycle PRD-XXX”Plugin: forgeplan-workflow
Explicit step-by-step forge cycle with 8 phases:
Phase 0: OBSERVE → forgeplan healthPhase 1: ROUTE → determine depthPhase 2: SPRINT → plan wavesPhase 3: BUILD → implementPhase 4: AUDIT → adversarial reviewPhase 5: FIXES → fix HIGH/CRITICALPhase 6: EVIDENCE → create + scorePhase 7: COMMIT → git + PR/forge-audit
Section titled “/forge-audit”Plugin: forgeplan-workflow
Multi-expert code audit with Forgeplan methodology integration.
Development Commands (dev-toolkit)
Section titled “Development Commands (dev-toolkit)”/audit
Section titled “/audit”Plugin: dev-toolkit
Launches 4 parallel expert agents:
- Logic — correctness, edge cases, race conditions
- Architecture — SOLID, coupling, DRY, naming
- Security — OWASP Top 10, injection, auth
- Tests — coverage gaps, test quality
Reports findings with severity: CRITICAL / HIGH / MEDIUM / LOW.
/sprint “goal”
Section titled “/sprint “goal””Plugin: dev-toolkit
Adaptive sprint planning that scales:
- 1 task → just do it
- 3-5 tasks → parallel execution
- 10+ tasks → wave-based sprint with dependency tracking
/recall
Section titled “/recall”Plugin: dev-toolkit
Restores session context from Hindsight memory — what you worked on, what was decided, what’s pending.
FPF Commands (fpf plugin)
Section titled “FPF Commands (fpf plugin)”/fpf “question”
Section titled “/fpf “question””Plugin: fpf
Auto-routes to the right reasoning mode based on your question.
/fpf decompose “system”
Section titled “/fpf decompose “system””Break a system into bounded contexts with roles, interfaces, and responsibilities.
/fpf decompose our authentication system/fpf evaluate “options”
Section titled “/fpf evaluate “options””Compare alternatives with F-G-R scoring and evidence assessment.
/fpf evaluate React vs Vue vs Svelte for our SPA/fpf reason “problem”
Section titled “/fpf reason “problem””Structured ADI reasoning: 3+ hypotheses → predictions → evidence check.
/fpf reason why our API response times degradedOrchestra Commands (forgeplan-orchestra plugin)
Section titled “Orchestra Commands (forgeplan-orchestra plugin)”/session
Section titled “/session”Plugin: forgeplan-orchestra
Full context restore: Hindsight memory + Forgeplan health + Orchestra tasks.
Plugin: forgeplan-orchestra
Bidirectional sync between Forgeplan artifacts and Orchestra tasks.
UX Commands (laws-of-ux plugin)
Section titled “UX Commands (laws-of-ux plugin)”/ux-review
Section titled “/ux-review”Plugin: laws-of-ux
Reviews frontend code against 30 UX psychology laws. Reports violations with severity and fix suggestions.
/ux-law “law name”
Section titled “/ux-law “law name””Look up a specific UX law with examples, violations, and best practices.
Research & Build Commands (dev-toolkit)
Section titled “Research & Build Commands (dev-toolkit)”/research “topic”
Section titled “/research “topic””Plugin: dev-toolkit
Quick research — study a topic, find patterns, understand how something works. Uses Explore agents.
/deep-research “topic”
Section titled “/deep-research “topic””Plugin: dev-toolkit
Deep multi-agent research (4-7 agents). Writes reports to docs/research/. Use before major work.
/build “research-dir”
Section titled “/build “research-dir””Plugin: dev-toolkit
Implement from existing research reports. Reads research output and creates implementation plan.
/synthesize “dir1” “dir2”
Section titled “/synthesize “dir1” “dir2””Plugin: dev-toolkit
Combine multiple research reports into a unified plan. Useful when you researched several topics and need one roadmap.
/do “task”
Section titled “/do “task””Plugin: dev-toolkit
Universal task executor — takes any description and figures out the right approach.
/wave “description”
Section titled “/wave “description””Plugin: dev-toolkit
Quick wave-based execution from current chat context. No research phase — just plan and execute.
/write-doc “type” “topic”
Section titled “/write-doc “type” “topic””Plugin: dev-toolkit
Create structured documents: RFC, guide, report, ADR. Uses templates and project context.
/team-up
Section titled “/team-up”Plugin: dev-toolkit
Launch Agent Teams for parallel implementation. Best for tasks that span multiple domains (backend + frontend).
Proactive Suggestions
Section titled “Proactive Suggestions”Claude Code can suggest commands based on what you’re doing:
| You say | Suggested command |
|---|---|
| ”изучи”, “разберись”, “найди” | /research |
| ”глубоко изучи”, “ресерч” | /deep-research |
| ”напиши RFC”, “доку” | /write-doc |
| Complex multi-step task | /do or /sprint |
| ”ревью”, “проверь” | /audit |
| Backend + frontend work | /team-up |
| Has research reports | /build |
| ”объедини”, “roadmap” | /synthesize |
When to Use What
Section titled “When to Use What”| Situation | Command |
|---|---|
| Starting new feature | /forge "description" |
| Code review before PR | /audit |
| Complex multi-step task | /sprint "goal" |
| Architecture decision | /fpf evaluate "A vs B" |
| Understanding a system | /fpf decompose "system" |
| Debugging issue | /fpf reason "why X" |
| Restoring context | /recall or /session |
| Frontend quality | /ux-review |
Workflow Chains
Section titled “Workflow Chains”Research → Build: /fpf decompose → /forge "implement X"
Multi-step feature: /forge "feature" → /audit → fix → /sprint "remaining tasks"
Architecture decision: /fpf evaluate "options" → /forge "implement winner"