AI Agent & Automation Workflows

Walk every unresolved review thread on a PR, triage each one, reply with a rationale of whether or not the comment will be acted upon, make the code change if warranted, and mark the thread resolved. Use when the user asks to address only the open PR comments without re-running CI, respond to review feedback, resolve review threads, or clear bot comments on a PR.

Amend the last git commit with a useful, descriptive message. Use when you want to rewrite a commit message, amend commit description, or improve the last commit message.

Manage recurring tasks in the local Argus daemon via its HTTP API. Use when the user wants to schedule a task locally in argus, create a cron-driven task, fire X every weekday/hour/morning, set up a recurring agent that needs local filesystem access (logs in ~/.argus, local databases, dotfiles), list or update existing argus schedules, or run an argus schedule now. Distinct from /schedule, which creates remote cloud routines without local access.

Safely add a directory to Claude Code's Bash tool PATH. Use when a binary runs in your terminal but is "command not found" inside Claude Code's Bash tool, when adding a directory to PATH for Claude Code, when configuring CLAUDEENVFILE, or when settings.json env.PATH did not work. Explains why env.PATH replaces rather than appends PATH and is not variable-expanded, why ~/.zshrc PATH exports are dropped, and the supported SessionStart hook fix.

Investigate flaky CI failures across multiple workflow runs to identify patterns, categorize root causes, and propose fixes. Use when asked to investigate CI failures, find flaky tests, diagnose test flakiness, or understand why CI is failing repeatedly.

Compare two competing PRs implementing the same feature, identify strengths of each, and combine the best parts into a single branch. Use when comparing PRs, combining implementations, cherry-picking between PRs, or resolving competing approaches.

Competing implementations with judge evaluation. Use when comparing multiple approaches, picking the best solution, or running a bake-off.

Set up cross-agent skill infrastructure so skills work across Claude Code, OpenAI Codex, and GitHub Copilot. Use for multi-agent compatibility, skill migration, cross-agent sync.

Use when hardening npm supply chain, pinning dependency versions, adding .npmrc security flags, or setting up Renovate and audit workflows. Locks down install-time scripts, registries, version ranges, and CI checks.

Explain any topic in the simplest possible terms using analogies and everyday examples. Use when the user says "explain like I'm 5", "eli5", "explain simply", "dumb it down", or "explain in simple terms".

Analyze a spec, PRD, or codebase to identify missing skills and agents, propose them, then write them. Use for skill gap analysis, agent gap analysis, bootstrapping agent capabilities from a spec, or equipping a project with skills.

Generate SVG logo alternatives with side-by-side comparison. Use when creating logos, branding, or project icons.

Orchestrate a stack of dependent Argus tasks that produce a chain of stacked PRs. Use when the user has a multi-milestone plan that should land as N PRs where each branches off the previous. Decomposes the plan into sub-tasks, wires basebranch + dependson, embeds the closing protocol in every sub-task prompt, and polls until the stack lands.

Summarize the prompts used in the current conversation as a numbered list. Use when the user wants to see the prompt chain, review their prompts, or share the sequence of prompts that produced a result.

Clean up merged and stale git branches, prune old local and remote branches safely. Use for cleaning up old branches or pruning stale remotes.

Re-review with fresh eyes. Zero regressions. Go slow and analyze everything. Use for a paranoid final review before merging.

Iteratively run /rereview, fix the findings, and loop until reviewers approve clean. Use for iterative automated review, when you want /rereview to loop until clean, or for a paranoid pre-merge review that auto-addresses every blocker.

Show skill usage statistics, charts, and suggestions for pruning unused skills. Use when the user asks "skill usage", "which skills do I use", "unused skills", "skill stats", or "skill suggestions".

End-of-session skill that captures session state to HANDOFF.md, appends to the session log, and prunes CLAUDE.md of stale/ephemeral content (the Resources section is protected from removal). Surfaces insights for Claude Code's native Auto Memory but does not write to the memory path itself. Use at the end of any work session for seamless handoffs. Trigger when the user says "/wrapup", "wrap up", "end of session", "session handoff", "save session state", "wrap this up", "let's wrap up", "close out this session".