Planning & Specification Workflows

PR-level merge gate. Audits the WHOLE pull request (not just the diff) against a merge-readiness contract: SPEC acceptance criteria met, all phases complete, docs updated per the doc map, Closes N present when issue-born, tests added at the right layer, CI green, branch off the default branch and independently mergeable, and the review-change axes clean (or consciously deferred to tracked issues). Verdict: merge-ready, or a ranked list of blockers — never merges, never edits. Triggers: "is this PR ready to merge", "audit the PR", "merge gate for N", "can this ship", "pre-merge review", "audit-pr".

Internal skill for the agentic-workflow repo. After editing one or more SKILL.md files, bumps their version: fields and updates every piece of documentation in the repo that must stay in sync: CHANGELOG.md, CHANGELOG.es.md, README.md, and README.es.md. Run before committing any skill change. On Claude Code and want hand-tuned per-skill model/effort tiers? Install the claude branch instead (npx skills add gtrabanco/agentic-workflowclaude) — see the README. This branch is model-agnostic: the skill inherits whatever model and effort your agent session is already using. Trigger phrases: "bump the skill version", "update the version", "update the changelog", "I just changed a skill", "log this change", "version bump".

Generate design documents from existing code. Use when the user wants to "generate design docs", "create documentation from code", "reverse engineer architecture", "extract design documents", "document existing system", "create docs from implementation", "generate screen design docs", "document UI components", or "create frontend design specs". Also use when the user says "設計書を生成する", "ドキュメントを作る", "コードから仕様書を作成", "画面設計書を生成", or "アーキテクチャドキュメントを書く". Analyzes source code and generates design documentation.

One-time workspace setup. Generates local context files (.claude/context/, git-ignored) that all workflows depend on — repo metadata, coding conventions, local repo paths, language configuration (discovered, not hard-coded), and provider configuration. Supports ADO, Jira, GitLab, GitHub, Zoho, and local-markdown as work item providers, plus GitHub CLI (gh-cli) and GitLab CLI (glab-cli) as no-MCP git providers. Run once before using /dev-workflow or /story-workflow.

Internal contract of the agentic-workflow pack: the machine envelope — the fixed JSON block an external orchestrator parses to route the next step (which command, which model tier) without a human reading the chat. Skills do NOT emit it on their own (except workflow-status, the sensor): a driver that wants it injects the canonical system-prompt snippet defined here and runs the repair loop on parse failure. Not a menu entry; this skill owns the schema, and docs/workflow/ORCHESTRATION.md documents the driver loop that consumes it.

Internal step of plan-feature: turn a feature-request issue into a scoped, sized, roadmap-mapped SPEC with Closes N traceability.

Internal step of plan-feature: interview a raw idea into a filled, sized SPEC, proactively asking only what the project's docs don't already answer.

Internal step of plan-feature: from a scoped SPEC, generate the planning artifact set scaled to the feature's size (XS/S → SPEC-only; M/L → full set with a hardening phase) and register the roadmap entry. Docs only — never code.

Read the whole paper, treat the bundle's marks as a unified editorial brief, emit a holistic minimal-diff plan plus a machine-readable companion.

Periodic, product-wide health check — the CTO's "where do we actually stand?" Sweeps the WHOLE codebase (not a diff, not a PR) across every applicable axis — correctness, architecture, security/cybersecurity, performance, tests, UX/UI, accessibility, SEO, brand, tech debt — PLUS process & docs (incomplete phases, aging issues, solvable known-issues, doc/workflow completeness) and roadmap coherence. Mines accumulated suggestions from feature docs. Output: a severity-ranked report and concrete PROPOSALS — issues to open, roadmap features to add or remove. NEVER auto-fixes; the user decides what to act on. Triggers: "audit the product", "full health check", "are we product-ready", "product-audit", "what's the state of the codebase", "CTO review", "tech-debt and roadmap sweep".

Internal brand & copy review pass of the agentic-workflow review pack — composed in-turn by review-change and product-audit; not a menu entry. Checks changed user-facing copy against the project's brand doc: voice, terminology, and honesty of claims — applies only to surfaces with user-facing text. Findings only; never edits code.

Platform-adaptive review orchestrator. Reviews the current change by running review-implementation (find → classify) AND invoking only the review skills that apply to this project and this change (code, security, verify, design, accessibility, brand, tech-debt, perf, SEO) — never the inapplicable ones (no accessibility/SEO/brand for a CLI, library, or infra change). Synthesizes one classified report plus an explicit manual-verification checklist. Findings only — never refactors. Triggers: "review this change", "full review before merge", "review-change", "run the right reviews for this", "what should I check before PR".

Internal tech-debt review pass of the agentic-workflow review pack — composed in-turn by review-change and product-audit; not a menu entry. Inventories shortcuts, duplication, stale abstractions and missing coverage on the changed surface, each with a re-trigger condition. Findings only; never edits code.

Use after implementing a feature or any code change to verify completeness and correctness before…

Internal SEO review pass of the agentic-workflow review pack — composed in-turn by review-change and product-audit; not a menu entry. Checks changed web pages/routes for indexability, metadata, and structured data — applies only to public web surfaces. Findings only; never edits code.

Internal run-it verification review pass of the agentic-workflow review pack — composed in-turn by review-change and product-audit; not a menu entry. Runs the project's gate and the changed behavior for real — commands, exit codes, observed output — instead of assuming; what cannot be executed goes to the manual checklist. Findings only; never edits code.

End-to-end autopilot: found the project if needed (one upfront interview — product, features, stack, architecture, quality bars, ops, autonomy, budget), create or adopt the complete roadmap, then ship it feature by feature through the full workflow (plan → execute → review → PR → merge gate) driven by /loop, with no further questions. Default: opens PRs, the human merges; --fullauto merges MERGE-READY PRs under non-negotiable safety floors. Triggers: "ship the roadmap", "build the whole app from the roadmap", "run the full workflow on autopilot", "ship-roadmap", "autopilot this project".