reflect-skills-from-me…
This skill should be used when the user asks to "reflect on skills from memory", "audit marketplace skills against memory",…
Resolves GitHub issues using isolated worktrees and test-driven development, then delegates PR creation to /github:create-pr so the quality gate and the /github:review-pr loop always run. This skill should be used when the user asks to "resolve an issue", "fix issue #123", or
$ npx -y skills add FradSer/dotclaude --skill resolve-issues --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/resolve-issuesContext preview
The summary Claude sees to decide when to auto-load this skill.
Resolves GitHub issues using isolated worktrees and test-driven development, then delegates PR creation to /github:create-pr so the quality gate and the /github:review-pr loop always run. This skill should be used when the user asks to "resolve an issue", "fix issue #123", or
name: resolve-issues allowed-tools: Bash(gh:*), Bash(git:*), EnterWorktree, ExitWorktree, Task, Skill description: Resolves GitHub issues using isolated worktrees and test-driven development, then delegates PR creation to /github:create-pr so the quality gate and the /github:review-pr loop always run. This skill should be used when the user asks to "resolve an issue", "fix issue #123", or needs to implement a solution for a specific GitHub ticket using a structured workflow. argument-hint: [issue number or description] user-invocable: true
Execute issue resolution workflow using isolated worktrees, TDD methodology, and agent collaboration.
Use isolated worktrees to avoid disrupting main development. Follow TDD cycle (red → green → refactor) with agent support. Reference issues in commits using auto-closing keywords. See `references/requirements.md` for protected PR workflow and commit standards.
**Goal**: Select target issue and prepare isolated development environment.
**Actions**: 1. Review open issues from context and select based on priority and `$ARGUMENTS` 2. Check existing worktrees to determine if reuse is possible 3. Use the EnterWorktree tool with a descriptive name (e.g., `fix-456-auth-redirect`) to create an isolated session 4. Rename the auto-generated branch to match conventions: run `git branch -m <type>/<issue>-<description>` (see `references/workflow-details.md` for naming) 5. Verify issue acceptance criteria and dependencies
**Goal**: Implement fix using test-driven development with agent collaboration.
**Actions**: 1. Plan implementation approach and assess architectural impact 2. Write failing tests that verify issue is resolved (RED phase) 3. Implement minimal code to make tests pass (GREEN phase) 4. Refactor while keeping tests green (REFACTOR phase) 5. Run quality validation commands to keep the TDD cycle honest (see `references/workflow-details.md` for project-specific checks). `/github:create-pr` re-runs the full gate in Phase 3 and is the authoritative pre-PR check.
**Goal**: Hand PR creation to `/github:create-pr` so the quality gate and the review loop run. Cleanup happens only after the merge, which may be many turns later.
**Actions**: 1. Push branch to remote with `git push -u origin <branch-name>` 2. **CRITICAL: Do NOT call `gh pr create` here.** Invoke `Skill("github:create-pr", "<issue reference>")` — e.g. `Skill("github:create-pr", "Closes #456")`. It is the plugin's only PR-creating path and owns the quality/security gate, the auto-closing-keyword linkage, the non-default-branch warning, and the mandatory `/github:review-pr` handoff. See `references/pr-creation-handoff.md` for the full contract. Creating the PR directly skips all of it.
3. **This skill does not resume here.** `/github:create-pr` reports the PR URL, and `/github:review-pr` then owns the PR for the rest of its life: a persistent Monitor spanning turns, the triage/fix/push rounds, and the merge decision it asks the user to make. Do NOT wait inline, do NOT re-report the URL, and do NOT run Phase 4 speculatively.
**Trigger**: The PR from Phase 3 has actually merged — normally a later turn, after `/github:review-pr` completed its merge decision. **`/github:review-pr`'s closeout now owns the post-merge cleanup** (worktree removal via `ExitWorktree action:"remove"`, switch to `main`, sync with origin), so this Phase runs only as a **fallback** when that cleanup was skipped: the user chose "Don't merge", an interrupt left the worktree behind, or this is a fresh session that cannot `ExitWorktree` the worktree created by an earlier session. Never assume the worktree is gone — verify first.
**Actions**: 1. Verify the merge with `gh pr view <PR#> --json state -q .state` returning `MERGED`; never assume. 2. Check `git worktree list` whether the issue worktree still exists. If `/github:review-pr` already removed it, skip straight to `git fetch --prune`. 3. If it persists: **CRITICAL: confirm still on the issue branch** before `ExitWorktree action:"remove"`. If checkout drifted onto `main`/`develop`, stop — removing would delete a long-lived branch. Remote head may already be gone; that is fine. 4. Use the ExitWorktree tool with action "remove" to clean up worktree and branch.
5. `git fetch --prune` to sync remote-tracking branches. 6. Document resolution and any follow-up tasks
A curated collection of 22 plugins for Claude Code, providing specialized agents, skills, and automation tools for development and productivity workflows.
This skill should be used when the user asks to "reflect on skills from memory", "audit marketplace skills against memory",…
Use acpx as a headless ACP CLI for agent-to-agent communication, always inside an isolated SubAgent. Use when running coding…
Delegates a self-contained task to a Google Gemini Managed Agent (Antigravity) running in a remote sandbox with code execution,…
Runs a deep-research query on Google Gemini's deep-research managed agent and returns a cited report. This skill should be used…
This skill should be used when the user asks to "understand a codebase", "get code context", "research a library", "explore a…
Execute this when the user requests code context for a repository, library, or any natural-language code/technology question…