agentic-coding

Batch-fix all open bugs in parallel, iterate on PR reviews until clean, then squash-merge each PR sequentially. Extends bug-blitz with automated review cycles and merge. Fetches bug-type tasks from Transit, creates worktrees, spawns parallel fix agents, runs /pr-review-fixer in a loop until no blockers/critical/major issues remain, then merges PRs one by one with squash-and-merge.

Batch-fix all open bugs in parallel using git worktrees. Fetches bug-type tasks in "idea" status from Transit, creates an isolated worktree per bug, and spawns parallel subagents each running the fix-bug workflow. Use when you want to tackle all open bugs at once, e.g. "fix all open bugs", "bug blitz", "resolve all Transit bugs".

Capture reusable cross-project technical knowledge into the user's Obsidian vault as a markdown note. Use whenever the user wants to save a learning, pattern, recipe, or how-to that would help future projects — phrases like "save this to my vault", "write this up as a note", "capture this for future reference", "document how X works", "remember this pattern", "add to my knowledge base", or any time after solving something non-obvious in their current repo. Examples are "how to set up an MCP server in a Swift app", "the GraphQL pagination pattern we used here", "how Transit wires up its dependency injection". Use this skill PROACTIVELY whenever the user finishes investigating something they'd plausibly want to reuse on a different project, even if they don't say "save it" explicitly — offer to capture it. Do NOT use for project-specific notes that only make sense inside one repo (those belong in that project's own notes), and do NOT use for daily/journal entries.

Catch Up on Branch Changes

Interactive Claude Code workshop that teaches features through hands-on exploration. Fetches the latest docs and changelog, organizes content into modules (basics, hooks, MCP, skills, subagents, etc.), and walks through each one step-by-step with explanations and live exercises in the current project. Supports version-based filtering to focus on what's new since a specific release (e.g., "everything from 2.1.90"). Use this skill whenever the user wants to learn Claude Code features, run a workshop, catch up on recent releases, explore what's new, get a guided tour of Claude Code capabilities, or says things like "teach me about Claude Code", "what's new in Claude Code", "workshop", or "walk me through the latest features".

Parallel code audit combining simplification and design critique analysis. Spawns two subagents (code-simplifier and design-critic) to independently review the current directory, consolidates findings into a prioritised report, and offers to create Transit tasks for actionable items. Use when you want a broad quality review of a codebase, e.g. "audit this code", "review codebase quality", "find improvements", "code audit".

Review and simplify code to reduce complexity and improve maintainability

Commits

Critical review of design documents, architecture proposals, and requirements

Analyze code for performance and efficiency improvements

Explain code changes or designs at three expertise levels (beginner, intermediate 5-10 yrs, expert). Use for: (1) PR/branch change explanations - understand what changed and why, (2) Design document validation - self-review by explaining at multiple levels to catch gaps or logic issues. Triggers: "explain this PR", "explain these changes", "validate this design", "explain like I'm a beginner", "walk me through this at different levels".

Systematic bug investigation, resolution, and documentation. Use when fixing bugs that need thorough analysis, test coverage, and a formal bugfix report. Applies systematic debugging methodology, creates regression tests, and generates a standardized report in specs/bugfixes/<bug-name>/. For complex bugs, spawns competing implementation agents (including alternative harnesses like Kiro) and selects the best solution. Triggers on requests like "fix this bug", "debug and document this issue", or when a bug needs both resolution and documentation.

Fix Go unit tests to comply to best practices. Use this when you're asked to modernise tests.

4. Make it so (implement all tasks)

4. Next task

Applies a modified Fagan Inspection methodology to systematically resolve persistent bugs and complex issues. Use when multiple previous fix attempts have failed repeatedly, when dealing with intricate system interactions, or when a methodical root cause analysis is needed. Do not use for simple troubleshooting. Triggers after multiple failed debugging attempts on the same complex issue.

Generate Claude Code permissions config from session history. Use when setting up autonomous mode, configuring .claude/settings.json, avoiding --dangerously-skip-permissions, or analyzing what permissions a project needs. Reads session logs to extract Bash commands and MCP tools actually used, then generates appropriate allow/deny rules.

Pull a PR from GitHub, run parallel review agents (code reuse, quality, efficiency, spec adherence) in read-only mode, surface unresolved comments, and write a self-contained HTML overview page to the repo root. Use whenever the user wants a read-only summary of a PR before deciding what to do — phrases like "overview of PR X", "summarise this PR", "what's the state of PR X", "give me an HTML overview", or any time they want a browseable artifact without touching the code.

Push a branch, open a change request, iterate on reviews until clean, then squash-merge. Runs /pr-review-fixer in a loop until no blockers/critical/major issues remain, rebases onto latest origin/main, and squash-merges. Works on both GitHub (gh) and GitLab (glab) — it detects the forge from the git remote. Works with or without Transit tickets. Use when you want to shepherd a PR/MR from push to merge, e.g. "push and merge this", "get this merged", "review-fix-merge loop".

Fetch unresolved change-request review threads (both diff-anchored and CR-level), validate issues, and fix them. Also checks CI status and fixes failing tests, lint errors, and build issues. Works on both GitHub (gh) and GitLab (glab) — it detects the forge from the git remote. Use when reviewing and addressing PR/MR feedback. Filters out resolved threads, keeps only the last Claude review comment per thread (matching the `<!-- claude-local-review -->` sentinel from the local-review agent), validates issues, posts a review report as a CR comment, then fixes validated issues.

Pull a PR from GitHub, run parallel review agents (code reuse, quality, efficiency, spec adherence), apply local fixes, and write a self-contained HTML review page to the repo root. Use whenever the user wants to review a PR before merging or asks to "review PR <number>", "look at this PR", "give me an HTML review of PR X", or wants a browseable artifact for a GitHub pull request — even if they don't explicitly say "HTML".

Review unpushed commits before pushing to remote repository

Initialize Claude Code project settings with standard language-specific permissions. Use when setting up a new project for Claude Code or adding standard configuration to an existing project.

Search the user's Obsidian vault for cross-project technical knowledge previously captured by the /capture-knowledge skill, and apply what's there to the current task. Use this PROACTIVELY at the start of any non-trivial work in a framework, language, or domain the user has touched before — the vault may already hold the gotcha you're about to rediscover. Trigger on explicit asks like "check my vault", "do I have notes on X", "search my knowledge", "what did I write about Y", "recall how I did Z", "look that up in my notes". Examples that should trigger: "do we have anything about embedding MCP servers in Swift apps?", "search my vault for Go pagination patterns", "what did I learn about SwiftData rollback?", "I'm about to wire up CloudKit sync — anything saved about that?". Also use BEFORE diving into integration recipes, SDK wiring, platform quirks, build-system glue, or language-specific gotchas (Swift actor isolation, Go generics, SwiftData, MCP, CloudKit, GraphQL pagination, App Intents, Hummingbird, NWPathMonitor, etc.) — a quick vault check beats reinventing a wheel the user already documented. Do NOT use for questions answerable from the current repo alone (read the repo instead), questions whose answer doesn't depend on the user's prior project work (general programming knowledge, language syntax, public API references), or queries about user preferences (those live in memory / CLAUDE.md, not the vault).

Skill for preparing the project for a release

Manage hierarchical task lists using the rune CLI tool. Create, update, and organize tasks with phases, subtasks, status tracking, task dependencies, and work streams for multi-agent parallel execution.

Triage-first batch bug fixing with streaming review and merge. Extends blitz-merge with an upfront triage phase that prioritises open bugs, groups them into conflict-free batches, and assigns each the minimum model tier capable of fixing it (Claude Code sonnet/opus, Kiro and Codex gpt-5.6 luna/terra/sol). Reviews and merges each PR as soon as its fix lands instead of waiting for the whole batch. Use for "smart blitz", "triage and fix all bugs", "fix bugs by priority", "batch-fix with the cheapest model", or any request to blitz bugs with prioritisation or cost awareness.

Identify and clean up stale or superseded feature specs in a project's specs/ directory. Use whenever the user wants to prune, tidy, audit, consolidate, or declutter specs, mentions outdated/overruled/replaced/superseded specs or a spec that has a v2 or improved version, or asks which specs are still relevant — even if they don't explicitly say "cleanup".

Generate or regenerate specs/OVERVIEW.md with a tabular summary, status, and per-spec file listings

Initialize a new spec with requirements, design, and task planning. Orchestrates the entire spec-driven workflow from feature idea to actionable task list.

2. Create Feature Design Document

1. Requirement Gathering

Small Spec (Smolspec) - Lightweight specification for changes that carry no contested decisions, whatever their size

3. Create Task List

Build macOS 26+ SwiftUI forms using Liquid Glass design language. Use when creating preferences panes, inspectors, configuration screens, or settings views in SwiftUI for macOS 26. Provides Grid-based layout with right-aligned labels, correct control sizing, and correct glass modifier usage (glassEffect as background for containers, glassEffect in ZStack for rare tiles). Triggers on "build a form", "settings screen", "preferences view", "inspector panel", or any macOS SwiftUI form layout work.

Route Transit tickets to the appropriate workflow based on task type. Use when a user mentions a T-[number] ticket reference and wants to work on it. Fetches task details and dispatches to the correct skill (fix-bug, creating-spec, planning mode, or asks for clarification).

Evaluate and improve user experience of interfaces (CLI, web, mobile)