Loading...
Loading...
Found 3,129 Skills
Use version control as a craft — atomic commits, buildable history, useful PRs, bisect-friendly main, recoverable mistakes. Use this skill whenever the task involves writing commits or PRs, choosing a branching model, deciding rebase vs. merge, recovering from a force-push or accidentally-committed secret, debugging a regression with `git bisect`, structuring a long change as a series of small reviewable steps, or judging whether a repo's history is readable. Use it especially when reviewing commit messages, PR descriptions, branching strategies, or merge policies. Built on Tim Pope and Chris Beams on commit messages, Paul Hammant on trunk-based development, Vincent Driessen on GitFlow (and his 2020 note retiring it for SaaS), Linus Torvalds on never rebasing public commits, and the Google Engineering Practices CL guide.
Use when upgrading a public GitHub repository README, repo metadata, docs map, or first-impression brand surface to match the Zonic/Evensong proof-first standard.
Autonomous NeMo-RL research agent workflow for directed hypothesis testing and open-ended discovery. Guides agents through the full experiment lifecycle: understanding recipes and environments, wiring RL or NeMo-gym runs, launching reproducible baselines and iterations, analyzing results, preserving human oversight, and using git plus TSV logs as the research ledger. Do NOT use for: bug fixes, code review, documentation, refactoring, dependency updates, or single-file changes.
Front door of the SDD flow. When the user asks for a feature or bug fix ("implementa…", "agrega…", "arregla…", "add…", "build…", "fix…"), FIRST scan the request + codebase context for load-bearing ambiguity and ask a few targeted clarifying questions (selectable options, recommended first) BEFORE writing any spec or code — then hand a well-formed goal to /opsx:propose (small) or the sdd-feature-flow workflow (large). Skip the questions when the request is already unambiguous. Modeled on GitHub Spec-Kit's /clarify.
Operate the Trylle git platform with the official `try` CLI. Use when an agent needs to authenticate to Trylle; inspect or manage repositories, pull requests, issues, stacked PRs, CI, actions secrets or variables, organizations, profiles, SSH keys, automations, or bots; print Trylle web URLs; call an uncovered public API endpoint; or use Trylle's AI, explain, draft, and diff commands.
Review a GitHub pull request using Alan's GitHub MCP tools
Use when implementing a GitHub issue end-to-end — MANDATORY for all dev tasks; spawns a repo worker devbox through speckit phases.
Dispatch implementation tasks to agent teammates in git worktrees. Triggers: 'delegate', 'dispatch tasks', 'assign work', or /delegate. Spawns teammates, creates worktrees, monitors progress. Supports --fixes flag. Do NOT use for single-file changes or polish-track refactors.
Restore and read workflow state after a context break — re-inject workflow phase, task progress, and behavioral guidance into the current session, reconcile state against git reality, and verify whether a workflow exists. Use when the user says 'resume', 'rehydrate', 'where were we', or runs /rehydrate, or when the agent has drifted after context compaction. Do NOT use for saving or mutating state (that is /checkpoint).
Review a PR or git scope with configurable anti-slop, custom-check, and change-boundary analysis.
Run a groomed `ready` GitHub issue through the whole build span unattended — worktree, implement, commit, review, fix, QA plan, and open PR — so one issue reaches a reviewable PR with no human at the keyboard. Use when the user says "afkkit", "run issue #N unattended", "work the ready issues while I'm away", "autopilot this issue", "take this issue to a PR without me", or wants the middle of the kit workflow driven end to end on its own.
Patterns for invoking the GitHub CLI (gh) from agents. Covers structured output, pagination, repo targeting, search vs list, gh api fallback.