autopilot

Plugin autopilot from marketplace code-assistants in awinogradov/code-assistants. Commands read from the repo's manifest — 1 month ago.

Skills (28)

• ascii-schemas

  Generate ASCII schemas, diagrams, and UI wireframes using Unicode box-drawing characters (wiretext conventions). Use when creating architecture diagrams, entity-relationship models, database schemas, flow charts, deployment topologies, sequence diagrams, data flow visualizations, UI wireframes, scre

• ask-codex

  Delegate a code analysis, refactoring, or automated-editing task to the OpenAI Codex CLI (codex exec / codex resume), then critically evaluate its output as a peer AI. Use when the user asks to run Codex, references OpenAI Codex, or wants a second model's take on the code.

• ask-gemini

  Delegate a code analysis, refactoring, or automated-editing task to the Google Gemini CLI (gemini -p / gemini --resume), then critically evaluate its output as a peer AI. Use when the user asks to run Gemini, references the Gemini CLI, or wants a second model's take on the code.

• branch-create

  Create and checkout a git branch following repository naming conventions with GitHub issue integration. Use when creating branches, or when invoked from other skills.

• commits-create

  Analyze staged changes and create conventional commits with intelligent grouping. Use when creating commits, or when invoked from other skills.

• commits-restructure

  Restructure messy draft commits into proper conventional commits

• dependabot-resolve

  Review and merge dependabot PRs safely using gh CLI. Process one-by-one to avoid package-lock.json conflicts

• explore

  Map this repository broadly, write a durable context brief to .claude/context/brief.md, then take surgical fixes one at a time with no plan, branch, or PR machinery. Use when you have an area rather than a task — "help me understand this repo", "explore the refactoring flow" — and want the codebase

• gather-context

  Resolve task intent, reuse validated evidence, and acquire missing planning context and emit a Context Map. Use when plan or run needs issue, standards, branch, and codebase context without loading raw documents into the parent conversation.

• issue-create

  Create a GitHub issue with a structured body (Context, What, Why, Scope, Solution) and curated labels via the gh CLI. Use when filing new issues, or when invoked from other skills.

• issue-run

  List recent open GitHub or Linear issues, pick one, and start autopilot on it via the run skill. Use to go from browsing issues to a running autopilot session in one step.

• linear-create

  Create a Linear issue with a structured body (Context, What, Why, Scope, Solution) and wizard-selected status, label, and assignee via the Linear MCP. Use when filing a Linear ticket on a linear-tracked project.

• linear-plan

  Plan a Linear issue exactly as the plan skill does, then store the finished plan in that issue's description — refreshing a rough ticket title along the way — so it outlives the session. Storing is unconditional and never gated.

• linear-run

  Run any Linear issue end to end. Executes a valid stored plan verbatim when one exists; otherwise drafts and implements a fresh plan without a human approval gate.

• pdf-create

  Generate a beautiful, brand-themed, multi-page PDF — report, research doc, six-pager, or playbook — from structured content using a bundled @react-pdf/renderer pipeline (direct rendering, no headless browser). Use when the user asks to create, generate, build, export, or design a PDF, report, whitep

• plan

  Perform deep analysis of the codebase, recent changes, and the requested task. Create a validated implementation plan

• pr-answer

  Answer a user comment on a PR review and update review state if needed

• pr-create

  Create a pull request with validated title and description following repository conventions. Use when creating PRs, or when invoked from other skills.

• pr-monitor

  Monitor a PR for CI check status, review feedback, and merge conflicts until it is ready for human review, approved, or merged. A packaged watcher waits for the pull request and wakes the agent only for an actionable event, so the model stays dormant through ordinary pending CI. Fixes CI failures, r

• pr-resolve

  Address PR review comments by analyzing feedback, making code fixes, and replying to reviewers. Use when resolving review feedback on a pull request.

• pr-review

  Review a pull request and provide constructive feedback with structured verdict. Used by awinogradov/code-review-action

• pr-update

  Update an existing pull request's title and description based on current branch commits. Use when PR needs to be refreshed after new commits or when asked to update PR.

• pr-validate

  Validate a PR title and branch name against repository contributing guidelines

• preflight-check

  Validate git working state before committing, branching, or opening a PR. Detects wrong branch, invalid branch names, forbidden history-changing commands, stale merged branches, uncommitted changes, and out-of-date main.

• run-primed

  Run a tracked issue end to end in a session already primed by /autopilot:explore, consuming the SHA-validated context brief instead of the broad codebase pass. Requires .claude/context/brief.md and fails loudly when it is missing, malformed, stale, or from another revision — it never downgrades to /

• run

  Plan and implement, then either report verified no repository change or deliver a PR; a Linear-issue input also gets the finalized plan stored on its ticket before implementation

• shared-rules

  Canonical home for instruction blocks shared by several autopilot skills and agents — reference formatting (RFC-0001), AskUserQuestion formatting and content-preview contract, agent structured output, issue body grammar, Linear MCP access, peer CLI delegation, git history policy, PR title/body gramm

• todo-cleanup

  Scan codebase for TODO/FIXME comments, verify actuality, create GitHub issues, and update links

Subagents (9)

• analyze-pr-commits

  Analyze branch commits, diff, and the linked GitHub or Linear issue for PR context. Use when pr-create or pr-update needs pre-computed context without polluting parent conversation.

• analyze-staged-changes

  Categorize staged files and assess commit strategy. Use when commits-create needs pre-computed analysis without polluting parent conversation.

• digest-repo-standards

  Read a repository's own CODE_REVIEW.md (when present) or its README, docs/, rfc/, and principles/, plus CLAUDE.md, and return a bounded standards digest. Use when planning skills need the project's conventions without loading the full documents into parent context.

• digest-session-history

  Map task-relevant files and commits to the Entire sessions and checkpoints that produced them via the entire CLI. Use when planning skills need session history without raw transcripts in parent context.

• resolve-alert-context

  Fetch GitHub code-scanning alert context via the code-scanning API. Use when plan/run resolve a code-scanning-alert input without polluting parent context.

• resolve-assignees

  Resolve candidate assignees for an issue from CODEOWNERS and Linear team members, with the current Linear user first. Use when a creation skill needs an assignee picklist without polluting parent context.

• resolve-issue-context

  Fetch issue context from GitHub (gh) or Linear (bundled GraphQL helper keyed by LINEAR_API_KEY) and optionally auto-assign the current user (idempotent, opt-in via caller flag). Use when commands need structured issue data without polluting parent context.

• scan-and-analyze-todos

  Scan codebase for TODO/FIXME comments and analyze their GitHub or Linear issue status. Use when todo-cleanup needs scan + analysis without polluting parent context.

• search-codebase-todos

  Search for TODOs and GitHub/Linear issue references in the codebase. Use when plan command needs TODO search in parallel with other context agents.

MCP servers (1)

• wiretext

  runs: npx -y @wiretext/[email protected]