attckdigital/smith/smith-help

Claude Code and agent skill by attckdigital/smith

Trust tier 2
  • community
LLM review pending
Source check pending

Checking that the SKILL.md file is still in the repository.

Central reference for all Smith commands — overview, detailed help per command, syntax, flags, and workflow context.

Install

npx vskill@latest install attckdigital/smith/smith-help

SKILL.md

Smith Help System

Display help for Smith commands. With no arguments, shows the full command overview. With a command name, shows detailed help including syntax, flags, examples, and workflow context.

Arguments: $ARGUMENTS

Behavior

This action is read-only — it does not modify any files. It should always be handled directly by the current session, never delegated to a sub-agent.

No Arguments — Full Overview

When invoked with no arguments, display this overview:

SMITH — Forge Your Workflow
FEATURE DEVELOPMENT
/smith-new Start a new feature end-to-end
/smith-specify Create or update a feature spec
/smith-clarify Ask clarification questions on a spec
/smith-plan Generate implementation plan from spec
/smith-tasks Break plan into ordered tasks
/smith-analyze Check consistency across spec/plan/tasks
/smith-implement Execute all tasks sequentially
/smith-build Autonomous build (tasks → implement → test → PR → merge)
/smith-checklist Generate quality checklist for spec/plan
/smith-bugfix Autonomous bugfix (branch → fix → test → PR → merge)
/smith-debug Diagnostic investigation (gather evidence → report → optional bugfix)
VAULT
/smith-vault Dashboard — recent sessions, agents, queue status
/smith-vault sessions List session logs
/smith-vault agents List sub-agent types and history
/smith-vault queue Show pending queue items
/smith-vault projects List all registered projects
/smith-vault status Vault health and routing metrics
QUEUE
/smith-queue add Add a task to the queue
/smith-queue list Show pending tasks by priority
/smith-queue process Pick and process a task (interactive, --next, --all, or <file>)
/smith-queue batch Alias for process --all
/smith-queue status Show tasks grouped by status
/smith-queue schedule Schedule a task for a specific time
/smith-queue history View completed and failed tasks
/smith-queue edit Modify a queued task
/smith-queue promote Change task from autonomous to interactive
/smith-queue demote Change task from interactive to autonomous
/smith-queue requeue Reset a failed task to pending
/smith-queue scheduler Manage the automated 2am batch scheduler
BANK
/smith-bank Deposit an idea from the current conversation
/smith-bank list Show all banked ideas (--priority, --system filters)
/smith-bank process <id> Withdraw idea and start /smith-new workflow
/smith-bank edit <id> Edit a banked idea's details
/smith-bank remove <id> Archive a banked idea
/smith-bank prioritize <id> --priority <level> Change priority
LEARNING
/smith-reflect Analyze completed workflow and extract lessons to the Ledger
/smith-reflect --last N Analyze last N workflows in batch
/smith-reflect --failure Learn from recent failures specifically
/smith-ledger Browse the Ledger dashboard
/smith-ledger list List all learned patterns by category
/smith-ledger search Search across all Ledger entries
/smith-ledger prune Remove stale low-confidence entries
/smith-ledger reset Archive and restart the Ledger
/smith-ledger reconcile Trigger Ledger self-maintenance (merge, prune, budget enforce)
/smith-ledger reconcile --dry-run Preview reconciliation without modifying files
/smith-ledger reconcile --force Bypass lock and threshold checks
/smith-ledger reconcile --status Check if reconciliation is needed
REPORTS
/smith-report Generate client-facing project report
PROJECT
/smith Bootstrap a new project with Smith workflow
/smith-constitution Create or update project constitution
/smith-taskstoissues Convert tasks.md to GitHub Issues
/smith-migrate-specs Migrate flat specs into system-based hierarchy
/smith-migrate-system-paths Add `paths:` frontmatter to existing system specs
SESSION
/smith-finish Commit, push, merge, update specs, clean workspace
/smith-sync Sweep & push team-shareable .smith/ artifacts to the default branch
HELP
/smith-help This overview
/smith-help <command> Detailed help for a specific command

After displaying the overview, also scan .claude/commands/ for any smith.*.md files not listed above. If found, append them under a "ADDITIONAL" section with their description from frontmatter. This ensures newly added commands appear in help even without explicit entries.

With Arguments — Detailed Help

When $ARGUMENTS contains a command name, display detailed help for that command.

Match the argument against known commands. Accept with or without the smith- prefix (e.g., both queue and smith-queue work). Also accept partial matches for sub-commands (e.g., queue process matches smith-queue process).


new
/smith-new — New Feature Workflow
The primary entry point for building features. Combines conversational
requirements gathering, spec generation, planning, a mandatory questions
gate, and autonomous build into a single flow.
USAGE
/smith-new <feature description>
/smith-new (prompts for description interactively)
NATURAL LANGUAGE TRIGGERS
"let's smith this"
"start a smith workflow"
"kick off a new feature for this"
"let's build this"
PHASES
1. Branch Safety Check current branch, pull main if needed
2. Requirements Conversational gathering or parse $ARGUMENTS
3. Spec & Branch Auto-detect system, create branch, generate spec.md
4. Plan Generate plan.md, research.md, data-model.md
5. Questions Gate Generate questions.md — MANDATORY STOP for user answers
6. Build Launch /smith-build for autonomous execution
SYSTEM DETECTION
Automatically determines primary_system and also_affects by analyzing
the feature description against .specify/systems/system-*/spec.md.
Feature spec is created at:
.specify/systems/<primary-system>/features/<NNN-short-name>/
CHAINS INTO
/smith-build (Phase 6 — autonomous from tasks through merged PR)
VAULT LOGGING
Logs invocation, system detection, spec creation, questions generated,
questions answered, and build handoff to the vault session log.
specify
/smith-specify — Feature Specification
Create or update a feature specification from a natural language description.
USAGE
/smith-specify <feature description>
BEHAVIOR
1. Generate branch short-name (2-4 words)
2. Find next feature number across branches and specs
3. Auto-detect primary system from .specify/systems/
4. Create feature branch and spec folder
5. Write spec.md with frontmatter (primary_system, also_affects, branch, status)
6. Generate quality checklist at <feature>/checklists/requirements.md
7. Validate spec (up to 3 iterations)
NEXT STEPS
/smith-clarify → /smith-plan
clarify
/smith-clarify — Spec Clarification
Identify underspecified areas in the current feature spec and resolve them
through up to 5 targeted questions.
USAGE
/smith-clarify
BEHAVIOR
Runs on the current feature branch. Scans the spec for ambiguities across
10 taxonomy categories. Presents one question at a time with recommended
answers. Updates the spec inline after each answer.
QUESTIONS FILE
Stored at <feature-folder>/questions.md (inside the spec folder, not specs/questions/)
NEXT STEP
/smith-plan
plan
/smith-plan — Implementation Planning
Generate a technical implementation plan from the feature spec.
USAGE
/smith-plan
ARTIFACTS PRODUCED
plan.md Architecture, file structure, tech decisions
research.md Resolved unknowns, dependency research
data-model.md Entity definitions and relationships
contracts/ API specifications
quickstart.md Integration scenarios
NEXT STEPS
/smith-tasks → /smith-analyze → /smith-implement
tasks
/smith-tasks — Task Generation
Break the implementation plan into ordered, executable tasks.
USAGE
/smith-tasks
OUTPUT
tasks.md with phases:
Phase 1: Setup (project initialization)
Phase 2: Foundational (blocking prerequisites)
Phase 3+: User Stories in priority order
Final: Polish & Cross-Cutting Concerns
NEXT STEPS
/smith-analyze → /smith-implement
analyze
/smith-analyze — Consistency Analysis
Non-destructive cross-artifact analysis across spec.md, plan.md, and tasks.md.
USAGE
/smith-analyze
CHECKS
- Spec ↔ plan alignment
- Plan ↔ tasks coverage
- Constitution compliance
- Duplicate or contradictory requirements
- Missing test coverage
OUTPUT
Analysis report with findings by severity (CRITICAL/HIGH/MEDIUM/LOW)
QUEUE INTEGRATION
After analysis, offers: "Implement now or queue for later?"
If queued, packages spec/plan/tasks into an autonomous queue entry.
implement
/smith-implement — Task Execution
Execute all tasks defined in tasks.md sequentially.
USAGE
/smith-implement
BEHAVIOR
Reads tasks.md, executes each task in order, marks completed with [X],
runs tests after each task. Logs per-task status to vault.
build
/smith-build — Autonomous Build Pipeline
Full build from tasks through merged PR. Runs without user interaction.
USAGE
/smith-build (normal — on a feature branch with spec/plan)
/smith-build (recovery — detects state and resumes)
PHASES
0. Context Discovery Load spec, plan, questions, tasks
1. Task Generation Generate tasks.md if missing
2. Implementation Execute tasks phase-by-phase in subagents
3. Testing Unit tests + Playwright E2E (if frontend changed)
4. Spec Updates Update system specs in .specify/systems/ and specs/
5. Commit & PR Stage, commit, push, create PR, squash-merge
6. Service Rebuild Docker rebuild affected services, health check
7. Release Notes Generate release.md, commit to main
RECOVERY
Detects current state and resumes from the right phase:
No tasks.md → Phase 1
Tasks partially done → Phase 2
All done, uncommitted → Phase 5
PR exists → Phase 5.3
PR merged → Phase 6
debug
/smith-debug — Diagnostic Investigation
Read-only workflow that systematically gathers evidence, identifies root
causes, and produces a structured debug report. Does not modify code.
The report feeds into /smith-bugfix if a fix is needed.
USAGE
/smith-debug <symptom description>
/smith-debug (prompts for details interactively)
NATURAL LANGUAGE TRIGGERS
"debug this", "help me debug...", "I'm getting this error...",
"why is X failing", "something is broken", "can you investigate..."
PHASES
1. Symptom Capture Extract or ask for: error, trigger, conditions, frequency
2. System Detection Map symptom to .specify/systems/<system>/
3. Automated Triage Parallel sub-agents: infra-health, log-scan,
dependency-trace, spec-cross-reference
4. Diagnosis Synthesize findings, rank hypotheses by evidence
5. Debug Report Write to .specify/systems/<system>/debug/debug-YYYY-MM-DD-<slug>.md
6. Decision Gate User chooses: fix (/smith-bugfix), investigate deeper, or close
SUB-AGENTS (launched in parallel based on symptom type)
infra-health docker ps, health-check.sh, docker stats, port checks
log-scan docker logs for affected + upstream services
dependency-trace map request path, check each hop end-to-end
spec-cross-ref prior debug reports, recent git changes, GitHub issues
REPORT STORAGE
.specify/systems/<primary-system>/debug/debug-YYYY-MM-DD-<slug>.md
CHAINS INTO
/smith-bugfix (if user selects "Fix it" at the decision gate)
bugfix

This is the start of the file. Read the full SKILL.md on GitHub.

Security

  • Security scanPassed, 100 of 100Pattern checks on the skill's files.
  • Model reviewNot runA model review is scheduled for the next rescan.

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

Repository stars
51
Repository forks
8
7-day trend score
20

Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/attckdigital/smith/smith-help/badge)](https://verified-skill.com/skills/attckdigital/smith/smith-help)

Something wrong with this skill or its page? Tell us.