Complete reference for every Vibeflow command: flags, modes, and outputs.
Deep-scans your codebase and learns your patterns, conventions, and architecture. Generates curated documentation in .vibeflow/ that persists and can be committed to git.
Usage /vibeflow:analyze [--fresh] [--scope <path>] [--interactive] [--satellite <url>]
Flags & modes (no flags) Incremental mode (default): re-analyzes only files changed since last run via git diff.
--fresh Full rebuild from scratch. Ignores existing .vibeflow/, re-analyzes everything.
--scope <path> Deep-dive into a specific module or directory (80%+ of files). Enriches global patterns with module-specific examples.
--interactive Adds a review checkpoint: presents patterns found, asks about false positives and missing patterns before saving.
--satellite <url> Analyzes an external dependency repo (design system, shared lib). Clones it, detects what your main repo uses, merges relevant patterns.
Output .vibeflow/index.md, conventions.md, patterns/, decisions.md
Interactive dialogue that transforms a vague idea into a clear, actionable PRD. Challenges assumptions, cuts scope aggressively, and forces decisions.
Usage /vibeflow:discover <idea or area>
Flags & modes Quick round Quick round: if your first response has total clarity (concrete problem, defined audience, closable scope), skips to 1–2 rounds then generates PRD.
Complete flow Complete flow: 3–5 discovery rounds with strategic questioning. Explores problem, audience, success criteria, scope, and trade-offs.
Output .vibeflow/prds/<slug>.md
Generates a technical spec with binary Definition of Done (3–7 pass/fail checks), scope, anti-scope, technical decisions, and risks. Grounded in your project's real patterns.
Usage /vibeflow:gen-spec <feature description or PRD path>
Flags & modes From PRD From PRD: pass a path to .vibeflow/prds/<slug>.md and it uses the PRD as basis for the spec.
From description From description: describe the feature directly and it generates the spec from scratch.
Auto-split Auto-split: if spec exceeds limits (>7 DoD checks or exceeds budget), automatically splits into numbered parts with dependency tracking.
Output .vibeflow/specs/<slug>.md (or <slug>-part-N.md if split)
Implements a feature from its spec with hard guardrails. Follows your patterns exactly, respects budget and anti-scope, runs tests, and self-verifies every DoD check.
Usage /vibeflow:implement <spec file or feature name>
Pipeline Phase 1 Find & validate spec: locates spec, validates required sections, checks multi-part dependencies.
Phase 2 Extract guardrails: DoD, scope, anti-scope, budget, and technical decisions become hard rules.
Phase 3 Load context: reads conventions, applicable patterns, index.md. Uses Pattern Resolution for smart matching.
Phase 4 Plan: identifies target files, verifies budget, maps DoD to implementation steps.
Phase 5 Implement: executes following patterns exactly. Minimum change principle. No refactoring outside scope.
Phase 6 Run tests: auto-detects test runner (npm test, pytest, cargo test, etc.). Max 2 fix attempts.
Phase 7 Self-verify DoD: checks each DoD item with evidence. Reports overall status.
Output Code changes within scope and budget, with DoD verification report.
Claude Code only. Budget is a hard limit. Anti-scope is sacred.
Generates a self-contained prompt for a coding agent in a separate session. The agent receives everything it needs: patterns, conventions, file paths, test commands. Zero external context.
Usage /vibeflow:prompt-pack <spec file or feature>
Flags & modes Agent-agnostic Agent-agnostic: works with Copilot, Cursor, Claude Code, or any coding agent.
Embedded patterns Embedded patterns: includes real code examples from your .vibeflow/patterns/ so the agent follows your conventions.
Validation Validation: rejects specs with >7 DoD checks or exceeding budget. Requires splitting first.
Output .vibeflow/prompt-packs/<slug>.md
Audits implementation against the spec on two dimensions: DoD compliance (every check pass/fail with evidence) and pattern compliance (conventions followed or deviations noted).
Usage /vibeflow:audit <spec file or feature>
Flags & modes PASS PASS: all DoD checks pass, patterns followed, tests pass.
PARTIAL PARTIAL: some checks pass, gaps identified with specific details.
FAIL FAIL: DoD not met or tests failing. Tests fail = automatic FAIL regardless of other checks.
Incremental fix On PARTIAL or FAIL, generates an incremental prompt pack covering only the gaps, so you can fix without re-reading the full spec.
Output .vibeflow/audits/<slug>-audit.md
Fast-track for small, well-defined tasks. Skips discovery, generates an ephemeral spec in memory, and outputs a ready-to-use prompt pack in one command.
Usage /vibeflow:quick <task description>
Flags & modes Budget ≤4 files ≤4 files budget: tighter than the standard ≤6. If the task needs more, it warns you to use the full pipeline.
Ephemeral spec Ephemeral spec: generated in memory only (not saved to disk). Goes straight to prompt pack.
When to use Use for bug fixes, small features, quick changes. Skip if the idea is vague or architecturally significant.
Output .vibeflow/prompt-packs/<slug>.md
Updates the .vibeflow/ knowledge base with corrections, new conventions, architectural decisions, or new patterns based on natural language feedback.
Usage /vibeflow:teach <feedback>
Flags & modes Pattern correction Pattern correction: updates an existing pattern doc with your correction. Survives incremental updates.
New convention New convention: adds to conventions.md (e.g., "we always use named exports").
Architecture decision Architectural decision: adds to decisions.md with context and discarded alternatives.
New pattern New pattern: creates a new .vibeflow/patterns/<name>.md file with standard structure.
Output Updates to .vibeflow/ docs (conventions.md, patterns/, decisions.md)
Compiles statistics from all audit reports. Shows pass/fail rates, most violated patterns, common DoD gaps, and quality trend over time.
Usage /vibeflow:stats
Flags & modes Metrics Verdicts breakdown (PASS/PARTIAL/FAIL percentages), DoD pass rate, top 3 most failing checks, top 3 most violated patterns.
Trend Quality trend: improving, stable, or degrading (requires ≥3 audits).
Output Chat output only (read-only, no files modified).
