dotfiles.vorpal

Turn a freeform work request into a standardized brief block that team-lead's Pre-flight HARD GATE consumes — collapsing goal verification to a single confirm. Parses the request, derives every brief field it can support, asks ONE batched AskUserQuestion round only for genuinely underdetermined fields, then emits the block verbatim and stops. Standalone operator-intake aid; writes no files, spawns nothing. Trigger: "brief", "create brief", "standardize this request".

Evolve agent definitions in agents/*.md via multi-agent self-review. Phase 0 includes a per-agent historical audit of recent Claude Code transcripts, history, agent memory, and stall signals (TeammateIdle, -r2 respawns, shutdown-rejection). Trigger: "evolve agents", "improve agents", "grow the team", "refine agents".

Audit coherence between agents/*.md and the skill ecosystem (skills/* + .claude/skills/*) across four dimensions, emit a Coherence Report + Remediation Manifest, and ROUTE fixes to evolve-agents/evolve-skills. REPORT-AND-ROUTE ONLY: despite the evolve- prefix it NEVER edits any agent/skill file and NEVER self-invokes the evolve-* skills. Run as a post-edit gate after standalone evolve-agents/evolve-skills edits (evolve-suite runs it automatically). Trigger: "evolve coherence", "audit coherence", "check agent/skill coherence", "cross-reference audit".

Review and improve the Claude Code configuration defined in the repo's Rust builders (src/user.rs, src/user/claude_code.rs) and related scripts via multi-agent self-review. Phase 0 includes a historical audit of recent Claude Code transcripts, history, and agent memory plus a git-history audit of the config sources. Evolves the config SOURCE (Rust builders), NOT live settings.json — for a one-off settings.json/permission/env edit use the bundled update-config skill (/config). Trigger: "evolve config", "improve config", "review config", "refine config sources".

Full team-spawning evolve-* orchestrator that collects LOCAL per-spawn + REMOTE aggregate Claude Code model metrics, categorizes every spawn into a task-tier → model-tier taxonomy, and applies evidence-grounded, operator-gated model-routing edits to team-lead.md. Trigger: "evolve model distribution", "improve model routing", "model distribution", "audit model tiers".

Review and improve skill definitions via parallel @staff-engineer agents. Evaluates 8 dimensions, enforces Content Gate and byte budget. Phase 0 includes a per-skill historical audit of recent Claude Code transcripts, history, and agent memory. Trigger: "evolve skills", "improve skills", "refine skills".

Author a single Architecture Decision Record at docs/adr/{NNNN}-{slug}.md. Loaded into the calling agent's context; the agent drafts the ADR per the format authority below. Trigger: "create ADR", "record this decision", "draft an architecture decision record", "log architectural decision".

Conduct a code review on a scoped artifact (PR, branch, uncommitted, staged, or files). Loaded into the calling agent's context; the calling agent applies the role-appropriate playbook — @staff-engineer runs the 6-dimension general review, @security-engineer runs the security-dimension review. The format authority for both roles' output lives here. NOT the bundled /code-review skill (which can edit the working tree via --fix); this project skill was renamed away from "code-review" to avoid that collision. Emits a structured verdict into the calling agent's context only — it does NOT post to the PR; to post findings as inline PR comments in the operator's voice use Skill(review-and-comment). Trigger: "code review", "review this PR", "review the diff", "security review of changes".

Author a single Product Requirements Document at docs/spec/{slug}.md. Loaded into the calling agent's context; the agent drafts the PRD per the format authority below. Trigger: "create PRD", "draft PRD", "write a product requirements document", "decompose this into a spec under docs/spec/", "write up requirements for", "scope this feature".

Scan code at a flexible <scope> and emit a REPORT-ONLY set of simplification / refactor opportunities, each grounded in one of the 12 code-philosophy principles in ~/.claude/agents/senior-engineer.md (repo: src/user/claude-code/agents/senior-engineer.md) (no new rubric). Idiomatic clarity first — fewer lines is the side effect, never the goal. Self-service scout for @senior-engineer; writes no files and applies no edits. NOT a formal review verdict (that is Skill(code-review-verdict)). Trigger: "simplify scout", "scout for simplifications", "find refactor opportunities", "scan for cleanup".

Author a single Technical Design Document at docs/tdd/{slug}.md. Loaded into the calling agent's context; the agent drafts the TDD per the format authority below. Trigger: "create TDD", "draft TDD", "produce a technical design document", "write the design for {feature}".

Read-on-demand reference home for team-wide doctrine relocated out of team-lead.md's system prompt — Runtime Discipline (R1-R7), Truth-First Debugging, Docs-Path Taxonomy, the Vorpal-managed tool inventory, Deep Collaboration mechanics, the recurring-pitfalls memory convention, the Shutdown Protocol (SP-1/SP-2), the Communication-Discipline rule-numbering convention, the Design-Complete Gate (per-pattern artifact/acceptance table, Design-source grammar), and the Fable-distilled completeness / Monitor-for-Orchestration watch-pattern heuristics. This is a reference index, NOT an invocable workflow — do not `Skill(team-doctrine)`. Read the specific `references/*.md` file a `CANONICAL:*-LOCAL` pointer in an agent file cites; there is no trigger phrase that should load this into context speculatively.

Author a single UX design spec at docs/ux/{slug}.md. Loaded into the calling agent's context; the agent drafts the spec per the format authority below. Trigger: "create UX spec", "draft UX spec", "author design spec", "design spec for the new CLI", "produce a design spec", "create UX design".

Verify a Docket issue's acceptance criteria against the implementation diff (static, evidence-based — NOT runtime app-behavior verification) and emit a structured verification report. Loaded into the calling agent's context; the calling agent (`@sdet`) drives verification, the skill enforces the format authority — verdict ladder, required sections, validation rules. No file written; the report is emitted into the agent's context. Trigger: "verify acceptance criteria", "verify Docket issue", "produce verification report" — NOT app/PR runtime checks (that is the bundled runtime `verify` skill, the name this skill was renamed away from to avoid collision).

Multi-agent consensus voting protocol. Standalone: spawns reviewers. Team: delegates to orchestrator. Computes weighted quorum via docket. Use for decisions needing structured validation. Trigger: "create vote", "vote on this", "consensus vote", "run a vote".

Conduct a peer design review on a UX spec, draft design, or user-facing surface and emit a structured review report across six UX dimensions. Loaded into the calling agent's context; the calling agent (`@ux-designer`) drives the review, the skill enforces the format authority — six dimensions, severity ladder, recommendation ladder, required sections, validation rules. No file written; the report is emitted into the agent's context. Invoke BEFORE implementation (spec/draft review). For post-implementation verification use Skill(design-qa). Trigger: "design review", "review UX spec", "peer design review", "review this design".

One-time bootstrap of docs/spec/ — spawns @staff-engineer agents in parallel to generate project specification files. Re-invocation prompts before overwriting existing specs; ongoing maintenance is handled by @staff-engineer during TDD/review work, not by this skill. Trigger on: "create specs", "generate specs", "bootstrap project specs", "create project specifications".

Read-on-demand reference home for team-wide doctrine relocated out of team-lead.md's system prompt — Runtime Discipline (R1-R7), Truth-First Debugging, Docs-Path Taxonomy, the Vorpal-managed tool inventory, Deep Collaboration mechanics, the recurring-pitfalls memory convention, the Shutdown Protocol (SP-1/SP-2), and the Communication-Discipline rule-numbering convention. This is a reference index, NOT an invocable workflow — do not `skill({ name: "team-doctrine" })`. Read the specific `references/*.md` file a `CANONICAL:*-LOCAL` pointer in an agent file cites; there is no trigger phrase that should load this into context speculatively.

Evolve agent definitions in agents/*.md via multi-agent self-review. Phase 0 includes a per-agent historical audit of recent Opencode SQLite telemetry, stats, agent memory, and unresolved dispatch signals. Trigger: "evolve agents", "improve agents", "grow the team", "refine agents".

Audit coherence between agents/*.md and the Opencode skill ecosystem (.opencode/skills/*) across four dimensions, return a Coherence Report + Remediation Manifest to the caller/orchestrator, and ROUTE fixes to evolve-agents/evolve-skills. REPORT-AND-ROUTE ONLY: despite the evolve- prefix it NEVER edits any agent/skill file and NEVER self-invokes the evolve-* skills. Run as a post-edit gate after standalone evolve-agents/evolve-skills edits (evolve-suite runs it automatically). Trigger: "evolve coherence", "audit coherence", "check agent/skill coherence", "cross-reference audit".

Review and improve the the legacy runtime configuration defined in the repo's Rust builders (src/user.rs, src/user/opencode.rs) via multi-agent self-review. Phase 0 includes a historical audit of recent Opencode SQLite telemetry, stats, and agent memory plus a git-history audit of the config sources. Trigger: "evolve config", "improve config", "review config", "refine the legacy runtime settings".

Full one-shot subagent-dispatch evolve-* orchestrator that collects Opencode SQLite/stats model telemetry, categorizes every spawn into a task-tier → model-tier taxonomy, and applies evidence-grounded, operator-gated model-routing edits to team-lead.md. Trigger: "evolve model distribution", "improve model routing", "model distribution", "audit model tiers".

Review and improve skill definitions via parallel @staff-engineer agents. Evaluates 8 dimensions, enforces Content Gate and byte budget. Phase 0 includes a per-skill historical audit of recent Opencode SQLite telemetry, stats, and agent memory. Trigger: "evolve skills", "improve skills", "refine skills".

Turn a freeform work request into a standardized brief block that team-lead's Pre-flight HARD GATE consumes — collapsing goal verification to a single confirm. Parses the request, derives every brief field it can support, asks ONE batched `question` round only for genuinely underdetermined fields, then emits the block verbatim and stops. Standalone operator-intake aid; writes no files, spawns nothing. Trigger: "brief", "create brief", "standardize this request".

Multi-agent consensus voting protocol. Standalone: spawns reviewers. Team: delegates to orchestrator. Computes weighted quorum via docket. Use for decisions needing structured validation. Trigger: "create vote", "vote on this", "consensus vote", "run a vote".

Author a single Architecture Decision Record at docs/tdd/adr/{NNNN}-{slug}.md. Loaded into the calling agent's context; the agent drafts the ADR per the format authority below. Trigger: "create ADR", "record this decision", "draft an architecture decision record", "log architectural decision".

Conduct a code review on a scoped artifact (PR, branch, uncommitted, staged, or files). Loaded into the calling agent's context; the calling agent applies the role-appropriate playbook — @staff-engineer runs the 6-dimension general review, @security-engineer runs the security-dimension review. The format authority for both roles' output lives here. NOT the bundled /code-review skill (which can edit the working tree via --fix); this project skill was renamed away from "code-review" to avoid that collision. Trigger: "code review", "review this PR", "review the diff", "security review of changes".

Post-implementation QA of a shipped user-facing surface against its `docs/ux/` spec; emits a structured QA report. Driven by `@ux-designer`; format authority for verdict/severity/sections. Invoke after the spec is implemented (not for spec review — that's `design-review`). Trigger: "design QA", "run design QA", "verify implementation against UX spec", "QA the shipped UX".

Conduct a peer design review on a UX spec, draft design, or user-facing surface and emit a structured review report across six UX dimensions. Loaded into the calling agent's context; the calling agent (`@ux-designer`) drives the review, the skill enforces the format authority — six dimensions, severity ladder, recommendation ladder, required sections, validation rules. No file written; the report is emitted into the agent's context. Invoke BEFORE implementation (spec/draft review). For post-implementation verification use Skill(design-qa). Trigger: "design review", "review UX spec", "peer design review", "review this design".

One-time bootstrap of docs/spec/ — spawns @staff-engineer agents in parallel to generate project specification files. Re-invocation prompts before overwriting existing specs; ongoing maintenance is handled by @staff-engineer during TDD/review work, not by this skill. Trigger on: "create specs", "generate specs", "bootstrap project specs", "create project specifications".

Author a single Product Requirements Document at docs/spec/{slug}.md. Loaded into the calling agent's context; the agent drafts the PRD per the format authority below. Trigger: "create PRD", "draft PRD", "write a product requirements document", "decompose this into a spec under docs/spec/", "write up requirements for", "scope this feature".

Author a single Technical Design Document at docs/tdd/{slug}.md. Loaded into the calling agent's context; the agent drafts the TDD per the format authority below. Trigger: "create TDD", "draft TDD", "produce a technical design document", "write the design for {feature}".

Author a single UX design spec at docs/ux/{slug}.md. Loaded into the calling agent's context; the agent drafts the spec per the format authority below. Trigger: "create UX spec", "draft UX spec", "author design spec", "design spec for the new CLI", "produce a design spec", "create UX design".

Verify a Docket issue's acceptance criteria against the implementation diff (static, evidence-based — NOT runtime app-behavior verification) and emit a structured verification report. Loaded into the calling agent's context; the calling agent (`@sdet`) drives verification, the skill enforces the format authority — verdict ladder, required sections, validation rules. No file written; the report is emitted into the agent's context. Trigger: "verify acceptance criteria", "verify Docket issue", "produce verification report" — NOT app/PR runtime checks (that is the bundled runtime `verify` skill, the name this skill was renamed away from to avoid collision).

Review a pull request across security and general-correctness lenses, then post each finding as an individual single-line inline comment on the PR — written in the operator's voice, under the operator's GitHub account, after per-item approval. Self-contained leaf skill: the calling agent runs the whole flow end-to-end; it does NOT spawn a team. Trigger: "review and comment", "review this PR and post comments", "inline review of a PR", "post my review comments on <PR>".

Scan code at a flexible <scope> and emit a REPORT-ONLY set of simplification / refactor opportunities, each grounded in one of the 12 code-philosophy principles in ~/.config/opencode/agents/senior-engineer.md (repo: src/user/opencode/agents/senior-engineer.md) (no new rubric). Idiomatic clarity first — fewer lines is the side effect, never the goal. Self-service scout for @senior-engineer; writes no files and applies no edits. NOT a formal review verdict (that is Skill(code-review-verdict)). Trigger: "simplify scout", "scout for simplifications", "find refactor opportunities", "scan for cleanup".

Audit coherence between Codex agent/persona sources and the skill ecosystem (.codex/skills/* + src/user/codex/skills/*) across four dimensions, emit a Coherence Report + Remediation Manifest, and ROUTE fixes to evolve-agents/evolve-skills for Codex source targets or operator review for project-local .codex/skills findings. REPORT-AND-ROUTE ONLY: despite the evolve- prefix it NEVER edits any agent/skill file and NEVER self-invokes the evolve-* skills. Run as a post-edit gate after standalone evolve-agents/evolve-skills edits (evolve-suite runs it automatically). Trigger: "evolve coherence", "audit coherence", "check agent/skill coherence", "cross-reference audit".

Author a single Architecture Decision Record at docs/tdd/adr/{NNNN}-{slug}.md. Loaded into the calling agent's context; the agent drafts the ADR per the format authority below. Trigger: "create ADR", "record this decision", "draft an architecture decision record", "log architectural decision"

Intake-only: turn a freeform work request into a standardized brief block for team-lead's Pre-flight HARD GATE; never execute, implement, fix, or edit the requested work. Derives supported fields, asks ONE batched AskUserQuestion round only for routing-relevant unknowns, then emits the block and stops. Trigger: "brief", "create brief", "standardize this request", "intake brief".

Conduct a code review on a scoped artifact (PR, branch, uncommitted, staged, or files). Loaded into the calling agent's context; the calling agent applies the role-appropriate playbook — @staff-engineer runs the 6-dimension general review, @security-engineer runs the security-dimension review. The format authority for both roles' output lives here. NOT the bundled /code-review skill (which can edit the working tree via --fix); this project skill was renamed away from "code-review" to avoid that collision. Trigger: "code review", "review this PR", "review the diff", "security review of changes".