orc plugin

• /orc:address[--auto[=guided|full]] [<pr-number>] [--repo <name>] [--repos a,b] (omitted = cu

  Answer reviewer comments on YOUR open PR — categorize unresolved threads, push fixes, and post replies. Workspace-aware across linked PRs.

• /orc:adr[<title>] [--supersede NNNN]

  Author an Architecture Decision Record (ADR) — produce a numbered docs/adr/NNNN-*.md following the orc:adr-writing template. Also handles supersession and status transitions on existing ADRs.

• /orc:ci[--auto[=guided|full]] [<pr-number>|<branch>] [--run <id>] [--watch]

  Diagnose and fix a red CI run — dispatches orc-ci-investigator, then routes on its verdict: auto-fix via orc-code-fixer, flake re-run, infra escalation, or /orc:debug hand-off

• /orc:context[generate|show|refresh|add-rule] [\"<rule text>\"] [--section <name>]

  Write or refresh docs/agents/project-context.md, the binding rules file executor agents load before code.

• /orc:debug[--jira <KEY>] [--repos a,b | --repo a | --all-repos | --this-repo] <bug descrip

  Systematic root-cause investigation, then fix with TDD — never proposes a fix before finding the cause; resumable diagnosis + regression test in .orc/ state. Workspace-aware. For a quick one-off look without session state, prefer the bundled /debug.

• /orc:deps[audit|outdated|upgrade] [pkg | --all-safe] [--prod-only] [--repos a,b | --repo

  Audit, survey, and upgrade project dependencies — one-bump-per-commit loop, majors escalated, never auto-applied. Verbs: audit (default) | outdated | upgrade.

• /orc:env[up|status|down] [--containerize-app] [--rebuild] [--fresh] [--down-volumes] [--

  Provision a fast containerized dev environment for the current repo or workspace; stays up for reuse across QA runs. Verbs: up (default) | status | down.

• /orc:evidence[<TICKET-KEY>] [--target <name>] [--web <url>] [--driver chrome|agent-browser] [

  Collect browser evidence scoped to a ticket (or shared context) and deliver it — evidence packet, then upload to the ticket or keep local (always asks). For proving a ticket's behavior without a code diff/PR. Workspace-aware.

• /orc:explain[--qa-dir <dir>] [--target <name>] [--style <animate-style>] [--voice <kokoro-vo

  Build a narrated "what was made" walkthrough from a QA packet — golden-path demo clips (Playwright, overlays off) + animate motion graphics + local kokoro-js voice, assembled by orc-explain; one preview gate before render; delivered via the evidence-publish gates

• /orc:fan-out[--from-plan] [--max <n>] [--agent <name>] [--repos a,b | --repo a | --all-repos

  Parallel-dispatch N independent tasks that share no state — investigations, multi-PR review, multi-repo or bulk-doc work. Standalone parallelism primitive with no surrounding lifecycle. Workspace-aware.

• /orc:flow[--auto[=guided|full]] [--type=feature|bug|refactor|docs] [--rfc] [--verbose] [-

  End-to-end feature/bug/refactor pipeline (plan → start → implement → QA → ship → address → cleanup) with an interactive gate at every phase. Resumable via /orc:resume. --jira <KEY> links a ticket. Workspace-aware.

• /orc:jira<verb> [args] [--project KEY] [--type Story|Task|Bug|Epic|Sub-task] [--parent KE

  Manage Jira tickets via acli, and bind/unbind a ticket key to the current orc session — persists jiraTicket in .orc/orc.json + checkpoint.md so /orc:status, /orc:resume, and /orc:ship all see the link.

• /orc:jira-breakdown[<brief> | <prd-path> | --epic <KEY>] [--project <KEY>] [--team-size <n>] [--dry

  Turn a feature brief, PRD, or existing Epic into a full Jira hierarchy behind a preview gate

• /orc:plan[--auto[=guided|full]] [--grill] [--issues] [--jira <KEY>] [--repos a,b | --repo

  Plan a feature or refactor — produces a TDD-shaped plan, optionally design-grilled, and decomposes it into independently shippable issues. --jira <KEY> links a ticket. Workspace-aware.

• /orc:postmortem[<short-slug>] [--severity sev-1|sev-2|sev-3]

  Author a blameless incident postmortem — timeline, root causes, action items — filing each P0 action item as a tracker issue. Status reaches Final only when all P0 items close.

• /orc:prd[<title>] [--interview] [--from-jira <KEY>]

  Author a Product Requirements Document (PRD) — produce a numbered docs/prds/NNNN-*.md following the orc:prd-writing template. Optionally interview-driven and optionally seeded from a Jira ticket.

• /orc:qa[--auto[=guided|full]] [--target <name>] [--web <url>] [--no-web] [--no-env] [--

  Pre-PR quality gate — browser-driven QA for web changes with a mandatory evidence packet against a provisioned environment. No QA-passed claim without artifacts. Workspace-aware. For a quick behavioral check without the evidence packet, prefer the bundled /verify or /run.

• /orc:release[--dry-run] [--tag-only]

  Cut a semver release for the current repo — bump computed from conventional commits, evidence preview, tag + GitHub release

• /orc:resume[<session-id-or-branch>] [--phase <n>] [--repo <name>]

  Resume an interrupted multi-phase orc command from its last checkpoint. Reads .orc/orc.json for active sessions, picks one, jumps to the next pending phase. Workspace-aware.

• /orc:retro[--session <branch>] [--transcript <path>] [--propose] [<what to look back on>]

  Retrospective on a coding session — proposes changes to the agent's environment (navigation pointers, automated checks, coding standards, steering files, tool economy, information access), most severe first; applies accepted in-repo changes and files the rest through the tracker gate

• /orc:rfc[--grill] [--review <path>] [<title>]

  Author a system-design RFC — proposes a non-trivial change BEFORE implementation, surfaces alternatives, and invites critique. Differs from /orc:plan (design settled) and /orc:adr (decision recorded).

• /orc:scaffold<package-or-service-name> [--type=lib|service|app|cli]

  Bootstrap a new package, service, or app — scaffold with proper README, Diátaxis-shaped docs, an initial test, and a first commit. No business code, just the well-shaped shell.

• /orc:setup[--tracker github|jira|local]

  Run-once repo interview that writes the tracker layer (docs/agents/issue-tracker.md, triage-labels.md, domain.md) with recommended answers

• /orc:ship[--auto[=guided|full]] [--draft] [--base <branch>] [--verbose] [--max-loc <N>] [

  Finalize and open the PR — verifies tests, presents commit/branch/PR options, executes the chosen path, with a soft LOC size-budget gate. Last command before review; adds a Jira Resolves trailer when bound. Workspace-aware.

• /orc:stack-pr[--smart] [--base <branch>] [--draft] [--max-loc <N>] [--repos a,b | --repo a |

  Break a too-big branch into a stack of smaller chained PRs — commit-based by default (one PR per Conventional Commit), or --smart for messy branches. Detects gh-stack, falls back to plain gh. Workspace-aware.

• /orc:start[--auto[=guided|full]] [--worktree <path>] [--jira <KEY>] [--repos a,b | --repo

  Start a feature — set up an isolated worktree, draft the plan, and write the first failing test before any production code. --jira <KEY> links a ticket and informs the branch name. Workspace-aware.

• /orc:status[--all] [--branch <name>] [--repo <name>]

  Show all active and recently-completed orc workspaces from .orc/orc.json. Read-only — never modifies state. Workspace-aware.

• /orc:trd[<title>] [--from-prd NNNN] [--from-jira <KEY>]

  Author a Technical Requirements Document (TRD) — produce a numbered docs/trds/NNNN-*.md following the orc:trd-writing template. Optionally seeded from a prior PRD.

• /orc:triage[<issue/PR number, Jira key, or request> | --sentry <id|url> | <pasted report>]

  Intake for any report (issue, PR, Jira, Sentry, paste): check it is valid, then brief, file, park or reject it

• /orc:wayfinder[chart <loose idea> | work <map ref> [ticket ref]]

  Plan a chunk of work too big for one session as a shared map of decision tickets on the configured tracker; each session resolves exactly one decision

• @agent-orc:orc-ci-investigatorsonnet

  Investigator role — diagnoses a failing CI run from gh logs and returns a classified failure report (test, lint, build, flake, infra, drift) with a fix list; never edits or re-runs. Dispatched by /orc:ci, /orc:flow's CI gate and /orc:address.

• @agent-orc:orc-code-fixerhaiku

  Executor role — applies a decided list of code changes, runs the project's test suite and reports the diff. Used by /orc:address and wherever the changes are already decided.

• @agent-orc:orc-debug-investigatoropus

  Investigator role — root-cause investigation for hard bugs: reproduce, minimise, hypothesise, instrument, locate the line. Returns a written diagnosis and a regression-test recommendation; never fixes. Dispatched by /orc:debug and /orc:incident.

• @agent-orc:orc-env-provisionersonnet

  Executor role — provisions a reproducible dev environment for QA (compose, devcontainer or Dockerfile; minimal compose generated otherwise), healthcheck-gated, writes docker-env-state.json; host-mode fallback. Dispatched by /orc:env, /orc:qa, /orc:flow.

• @agent-orc:orc-implementersonnet

  Executor role — implements a slice list from a plan and failing tests: red-green-refactor, one commit per slice, suite green between. Default executor in /orc:flow Phase 5 and /orc:fan-out. Escalates on ambiguity, new dependencies, or three red attempts.

• @agent-orc:orc-jira-architectopus

  Investigator role — drafts a Jira backlog hierarchy (Epic micro-PRD, Stories, concurrency-sliced Tasks) per orc:jira-hierarchy and returns it as JSON; never creates tickets. Dispatched by /orc:jira-breakdown, /orc:flow and /orc:plan.

• @agent-orc:orc-pr-reviewersonnet

  Investigator role — reviews someone else's open GitHub PR and returns structured findings per orc:review-contract; never posts, never edits. Dispatched by /orc:code-review and /orc:fan-out.

• @agent-orc:orc-prd-analyzeropus

  Investigator role — analyses a PRD, spec or brief into structured requirements, ambiguities, missing edge cases and clarifying questions. Dispatched by /orc:flow triage and /orc:plan when the input reads like a brief.

• @agent-orc:orc-qa-validatorsonnet

  Executor role — drives a real browser via the agent-browser CLI to QA a running web app and writes the evidence packet (criterion screenshots, recording, console, HAR, steps.md, qa-manifest.json) to .orc/<branch>/files/qa/. Dispatched by /orc:qa for web changes.

• @agent-orc:orc-refactor-architectopus

  Investigator role — scans a codebase for refactor opportunities: coupling that hurts testability, duplicated decisions, drift from ADRs and the glossary. Dispatched by /orc:plan for refactor-shaped input and by orc:improve-codebase-architecture.

• @agent-orc:orc-reply-drafterhaiku

  Investigator role — drafts one short reply per PR review comment from the comment, the code and what changed; text only, never edits. Used by /orc:address.

• @agent-orc:orc-security-revieweropus

  Investigator role — reviews a diff or a whole repo for security vulnerabilities and returns orc:review-contract findings with an exploit scenario each; never edits. Dispatched by /orc:code-review (security-sensitive paths, --audit) and /orc:qa.

• @agent-orc:orc-stack-analyzeropus

  Investigator role — analyses a too-big branch and proposes a stack of smaller PRs as a JSON plan with a non-interactive rebase plan; never runs git mutations. Used by /orc:stack-pr --smart.

• @agent-orc:orc-test-authorsonnet

  Executor role — designs and writes test suites for new or changed behaviour: edge cases, error paths, fixtures. Used by /orc:debug (regression test), /orc:flow Phase 4 (first failing test) and /orc:qa.

• adr-writing

  Author and update Architecture Decision Records to document technical decisions, trade-offs, and alternatives in markdown format.

• agent-browser

  Provides a fast Rust-based CLI for browser automation using Chrome DevTools Protocol to navigate, scrape, and test web or Electron applications.

• architecture-patterns

  Implement Clean, Hexagonal, and Domain-Driven Design patterns to build maintainable, testable, and scalable backend systems.

• browser-qa

  Executes a standardized browser testing protocol using Playwright, agent-browser, or Claude-in-Chrome to validate web changes against acceptance criteria.

• callouts

  Formats structured emoji-header blockquotes for terminal conversations or GitHub-flavored markdown admonitions.

• caveman-pr

  Generates ultra-compressed, signal-only pull request titles and bodies following strict structural rules.

• caveman-review

  Generates ultra-compressed, one-line code review comments containing the location, problem, and fix.

• ci-routing

  Executes a standardized protocol for handling CI diagnosis verdicts including fixable, flake, infra, or debug-required scenarios.

• cleanup

  Removes workspace state, registry entries, git worktrees, and merged branches after completed or abandoned sessions.

• code-discovery

  Navigates and discovers codebases using a code graph to reduce token costs, with automatic fallback to grep and globbing.

• code-review

  Automates GitHub pull request reviews by posting inline comments, suggestion blocks, and computed verdicts based on finding severities.

• codebase-design

  Provides a shared vocabulary and principles for designing deep modules with small interfaces and high leverage.

• dependency-management

  Audit, survey, and upgrade project dependencies safely using ecosystem-specific tools and a strict one-bump-per-commit workflow.

• dispatching-parallel-agents

  Dispatch specialized subagents with isolated, self-contained briefs to solve independent problems concurrently.

• doc-writing

  Provides shared scaffolding, numbering conventions, and review workflows for technical design and specification documents.

• docker-best-practices

  Docker expert — image optimization, security hardening, multi-stage builds, orchestration, production deployment

• documentation-writing

  Write high-quality software documentation using the four-quadrant Diátaxis framework: tutorials, how-to guides, reference, and explanations.

• domain-modeling

  Maintain a project's ubiquitous language and decision records by managing GLOSSARY.md and Architecture Decision Records (ADRs).

• env-provisioning

  Provisions fast, reproducible development environments using Docker with healthcheck-gated booting and automated detection of existing configurations.

• error-handling-patterns

  Implements error handling strategies including exceptions, Result types, propagation, and graceful degradation across multiple programming languages.

• evidence-publish

  Delivers QA evidence packets to trackers like Jira or keeps them local after a mandatory user preview gate.

• executing-plans

  Execute a written implementation plan by reviewing it critically, performing tasks batch-by-batch, and stopping for review at checkpoints.

• explainer

  Builds narrated walkthrough videos for QA packets by combining scripts, demo footage, animated graphics, and AI narration.

• finishing-a-development-branch

  Guides the completion of a development branch by verifying tests and providing structured options for merging, creating pull requests, or discarding work.

• gates

  Defines standardized conventions for AskUserQuestion calls to manage user decision points through gate classes, previews, and autopilot modes.

• gh-cli

  Manage GitHub repositories, issues, pull requests, actions, and releases directly from the terminal using the gh command-line tool.

• git-advanced-workflows

  Manage complex repository histories using advanced techniques like rebasing, cherry-picking, bisecting, and worktrees.

• git-commit

  Generate standardized conventional commits by analyzing diffs and staging changes according to semantic specifications.

• grill-me

  Conducts a decision-by-decision interview to refine and sharpen plans or designs through critical questioning.

• grill-with-docs

  Conduct design grilling sessions that simultaneously generate documentation, glossaries, and architecture decision records.

• grilling

  Conducts round-based interviews to stress-test plans and design decisions by mapping them into a decision tree.

• handoff

  Summarizes the current conversation into a redacted document saved to the OS temporary directory for use by a new agent session.

• improve-codebase-architecture

  Identifies architectural friction and proposes refactoring opportunities to improve module depth, testability, and AI-navigability.

• incident

  Manages live production incident response by triaging evidence, choosing mitigation strategies, and maintaining a UTC timeline for postmortems.

• inline-review

  Converts structured findings into real GitHub pull request reviews with line-anchored inline comments and suggestion blocks.

• jira-cli

  Manage Jira work items from the terminal, including creating tickets, linking issues, performing JQL searches, and transitioning statuses.

• jira-hierarchy

  Enforces a standardized hierarchy and template structure for Jira Epics, Stories, and Tasks to ensure technical clarity and developer concurrency.

• map

  Acts as a router to recommend the appropriate orc command, skill, or workflow based on a given development situation.

• mobile-ios-design

  Design native iOS interfaces and build SwiftUI views following Apple's Human Interface Guidelines.

• nestjs-best-practices

  NestJS best practices and architecture patterns for production apps

+48

• PostToolUsecommand

  .*

• PreCompactcommand

  No description in the file.

• PreToolUsecommand

  Bash

• SessionStartcommand

  startup|resume|clear|compact, startup

• WorktreeCreatecommand

  No description in the file.

• WorktreeRemovecommand

  No description in the file.