jacobpevans-claude-code-plugins

Plugin jacobpevans-claude-code-plugins from marketplace cascade-code-general-misc-1 in joshuarweaver/cascade-code-general-misc-1. Commands read from the repo's manifest — 1 month ago.

Skills (63)

• goal

  Emit one goal statement for current work, capped under 4000 chars. Optional focus hint like /compact. Derives from recent pivots/plan/tasks if no arg. Use for 'what am I trying to do', seeding a session/subagent, or /handoff's goal half.

• handoff

  Emit a paste-ready two-part handoff: a /goal-capped Goal statement plus an unbounded Full prompt (cwd, reading list, rules, pitfalls, deliverables). Repo-optional. Use when forking to a new session/orchestrator, or wrap-up needs a resume prompt.

• replan

  Rebuild a plan from live ground truth (artifacts, tests, git/gh) when it no longer matches reality — items done, approach superseded, facts drifted. Use when a plan goes stale or resuming reveals contradictions. Never trusts the plan's own prose.

• resume

  Pick up unfinished work cold by re-deriving state from live git, gh, the plan file, and TaskList — never from remembered claims. Use at the start of a session continuing prior work. Verifies what's merged, open, committed, checked-off first.

• session-status

  Analyzes current session state, no cleanup. Full mode (default): plan+TaskList+history scan, emits a /handoff prompt. Mid mode (`/session-status mid`): fast 'done vs remaining' snapshot. Repo/PR state is optional; both modes run outside a repo.

• track-followups

  Record triaged follow-up work where it'll be seen: create the item in Vikunja or Zammad (incidents), deduplicating first and reporting the id. Use when session work outlives it or items were triaged elsewhere. Never opens a GitHub issue.

• wrap-up-docs

  Emit a paste-ready prompt for a weaker LLM to update public/private docs from this session's changes: per-item changelog, fixed routing, docs-only scope fence. Read-only, git-first evidence. Use at session end before delegating docs.

• wrap-up

  End-of-session handler: checks if the plan is complete. If complete: quick retrospective, forward follow-up prompt, and (in a repo) refresh + clean gone branches. If incomplete: skip cleanup, emit resume prompts. Cleanup needs a repo; rest doesn't.

• auto-maintain

  Autonomous maintenance orchestrator that continuously finds and dispatches work

• delegate-to-ai

  Route a task to the right model — a native Claude subagent, Codex, or local MLX — based on task type

• fast-subagent

  Use for any routine/checkable step — summarize, extract, classify, reformat, draft boilerplate, first pass over a file — every such step, all session. Sends to the router's fast-subagent role, which picks backend/fallback. Names no model ids.

• local-subagents

  Use for bulk reading, summarizing, classifying, extracting, drafting boilerplate, or a first pass over code — hand it to a local/cheap model via the router, not your context. Covers delegation timing, the model menu, bounding calls, refusals.

• multi-model-review

  Fan a plan/diff to independent model families — cloud coding-agent CLI, cloud reasoning-agent CLI, local model server — for adversarial review, plus each invocation's gotcha. Use for a second opinion, a suspect verdict, or an offline reviewer.

• openrouter-models

  Choose among the router's hosted models — discover current ids/prices from the public catalog, self-enforce a spend budget the router doesn't meter, respect the free-tier logging caveat, and request onboarding of a model not yet served.

• premium-agent-orchestration

  Use when the session runs a premium top-tier/SOTA model AND the work is independent pieces exceeding one context window, or insurance against an occasional runaway task in a large routine batch. Fans work to cheaper agents or local/free LLMs while the lead keeps judgment, merging, and verification.

• code-quality-standards

  Use when writing or reviewing code, choosing a logging format, or deciding between continuous monitoring and one-time tests in this estate

• codeql-permission-classification

  Permission requirements for GitHub Actions

• github-workflow-security-patterns

  Canonical security patterns for GitHub Actions workflows

• native-first

  Stack-agnostic playbook for the native, non-custom way to do something before writing a script or wrapper. Use when tempted to add a dependency or build a helper, or content-guards blocks a write. Climbs a fixed ladder; ends against silent fallback.

• validate-readme

  Validate all README files in the repository for required sections and installation code blocks, and score each one 0-6 on a quality rubric (purpose paragraph, quickstart, usage examples, license, length). Checks section presence using config from .readme-validator.yaml or sensible defaults. Badge UR

• git-flow-next

  Work with repositories using the git-flow-next branching model (default branch is `develop`, production branch is `main`). Covers branch creation, worktree setup, atomic commits, PR targeting, validation on `develop`, and mandatory end-of-session promotion to `main`.

• git-workflow-standards

  Use when managing branches, resolving merge conflicts, syncing with the default branch, or working with worktrees. Covers both trunk repos (main) and git-flow repos (develop) — see gh-cli-patterns (github-workflows) for default-branch detection.

• pre-commit-architecture

  Use when adding/editing .pre-commit-config.yaml, wiring pre-commit hooks into a repo, or scaffolding a new repo's lint/hook setup. Covers the canonical nix-devenv/dryvist-.github architecture, profiles, and consumer patterns.

• sync-main

  Update the repo's default branch from remote and merge it into the current branch, or all open PR branches. Despite the name (kept for continuity), this syncs whatever the repo's default branch actually is — main on trunk repos, develop on git-flow repos — and on git-flow repos also fast-forwards lo

• troubleshoot-precommit

  Troubleshoot pre-commit hook failures and auto-fixes

• troubleshoot-rebase

  Diagnose and recover from git rebase failures

• troubleshoot-worktree

  Troubleshoot git worktree, branch, and refname issues

• finalize-pr

  Automatically finalize pull requests for merge by resolving CodeQL violations, review threads, merge conflicts, and CI failures. Handles single PR (current branch or by number), all open PRs in the repo, or all open PRs across the org. Includes bot-authored PRs in all modes. argument-hint: "[PR_NUMB

• gh-cli-patterns

  Canonical reference for all gh CLI command shapes used by skills in this plugin. Defines the placeholder convention, default-branch (trunk vs git-flow) detection, allowed --json fields, GraphQL fallback rules, -f/-F/--raw-field flag semantics, the PR-readiness gate, code-scanning alert query, review

• github-actions-silent-failures

  Diagnose GitHub Actions workflows that fail silently — a reusable-workflow run with zero jobs/logs, a gh-aw import mis-resolving vs a repo named .github, self-hosted-runner traps. Use when a required check never reds or a run has no jobs.

• github-code-search

  Search real code across ~1M public GitHub repos via grep.app — use to see how others solved something, find real API/config-key usage, check if a pattern is idiomatic, or gather prior art. Works via the grep MCP server or a keyless HTTP fallback.

• issue-sweep

  Use when a tracker's open issues piled up, went stale, or stopped reflecting reality — maybe fixed, duplicated, vague, superseded — reconcile against actual code, group by root cause, turn into a plan. Any tracker. Use before a work session too.

• merge-pr

  Merge a PR into its base branch. Defaults to a merge commit; pass --squash or -s to squash instead. Falls back to squash when the repo disallows merge commits. Single PR by number or current branch. Refuses to squash/rebase into main on a git-flow repo — use /promote-release there instead. argument-

• pr-stacks

  Use when a change is too large for one PR, when the next unit of work depends on an unmerged PR, or when GitHub stacked pull requests / `gh stack` come up. The single source of truth for the PR-stack concept and the local pre-GitHub workflow: layering branches, keeping layers rebased, and landing a

• pr-standards

  Use when creating PRs, linking issues, managing PR comments, or creating GitHub issues

• pr-sweep

  Use when open PRs piled up in one repo or an owner's repos and you want them triaged and driven to zero in one pass — bot/teammate-authored included. Also for sweeps running many repos in parallel without racing approvals or flooding CI.

• promote-release

  Open and merge a develop-to-main promotion PR on a git-flow repo. Merge commit only, never squash or rebase — main accepts merge commits only. Every merge to main triggers release-please, which takes over tagging and release notes from there. Refuses on trunk repos (default branch main) — use /merge

• prune-branches

  Delete branches and worktrees with nothing useful — zero commits ahead of default, or a matched merged PR — local and remote, single-repo or workspace-wide. Run after refresh-repo syncs the default branch.

• rebase-pr

  Local rebase-merge workflow for pull requests with signed commits, pushed directly to the PR's base branch. Refuses on git-flow repos when the base is main — direct pushes to main are banned there; use /promote-release.

• refresh-repo

  Check PR merge readiness and sync the local repo with its default branch. See prune-branches (github-workflows) for stale branch and worktree cleanup.

• resolve-pr-threads

  Orchestrates resolution of GitHub PR review threads AND reads recent non-thread PR comments (top-level + review bodies) by grouping related feedback, processing each group sequentially inline with superpowers:receiving-code-review, and resolving threads via GraphQL. Use when you need to batch-proces

• shape-issues

  Shape raw ideas into actionable GitHub Issues using Shape Up methodology

• shared-workflow-org-refs

  Use when creating or editing GitHub Actions workflows that call reusable workflows (uses: OWNER/repo/.github/workflows/...) — org owner references must be the literal current org, and shared-CI homes are under dryvist.

• ship

  Commit, push, create PR(s), and auto-finalize — full automation pipeline. Handles uncommitted changes and recently created PRs.

• trigger-ai-reviews

  Use when asked to "trigger AI reviews", "request AI re-reviews", "get Claude/Gemini/Copilot to review my PR", "re-review this PR", or trigger review-request comments on a PR. Triggers all three AI reviewers by default, or specific ones when named. argument-hint: "[PR_NUMBER] [claude|gemini|copilot|a

• dell-idrac-bmc-ops

  Bring a Dell PowerEdge BMC (iDRAC) onto the network headlessly, standardize settings, stage firmware, drive an unattended install — Redfish/racadm gotchas, identity-before-power, phantom-drive from stale firmware. Use bringing up/diagnosing a BMC.

• homelab-runbooks

  Use when powering on a sleeping DR node, converging DNS records for a new ingress vhost, or bringing up OpenBao/Terrakube JWT identity — homelab operational runbooks for power management, DNS, and secrets-engine bring-up.

• llm-router-ops

  Operate a self-hosted OpenAI-compatible LLM router — client-wiring per client, context-window ad gotcha, env-vs-persisted-config gotcha, why an unauth health probe should 401. Use wiring a client, adding a backend, diagnosing misbehavior.

• proxmox-cluster-ops

  Operate a Proxmox VE cluster safely — read-only inspection with pvesh/pct/qm/pvecm instead of hand-editing a live guest, quorum-respecting node updates, joining a new node. Use for cluster/guest inspection, rolling updates, or adding a node.

• pxe-netboot

  Reliable remote OS install over PXE when virtual-media/KVM is broken — proxyDHCP+TFTP via dnsmasq, kernel/initrd over HTTP, custom iPXE with embedded autoexec. Use when a BMC's virtual CD is unreliable or a headless install needs netboot.

• terrakube-ops

  Operate self-hosted Terrakube (remote OpenTofu/Terraform plan/apply, state, locking) — login/plan/apply flow, why targeted apply is dangerous, lock recovery, offline-mirror gotcha, token rotation. Use for a plan/apply, stuck lock, new workspace.

• workstation-offbox-backup

  Keep a data-heavy app's local database small while preserving full history off-box, readable without restore. Splits by shape — text to log/search, media to snapshotted storage — pulls data, never a push script. Use when db/disk footprint grows.

• zfs-resumable-transfers

  Always send large ZFS datasets as resumable (zfs receive -s / zfs send -t), never plain zfs send | zfs receive -F. An interrupted transfer discards everything sent, turning a blip into hours of retransfer. Use for non-trivial zfs send/receive.

• orchestrate-infra

  Master orchestrator for cross-repo infrastructure with dependency graph dispatch

• sync-inventory

  Verify or refresh the OpenTofu-to-Ansible inventory distribution (RustFS-published; apply is the publish boundary)

• test-e2e

  End-to-end infrastructure pipeline validation across Terraform and Ansible repos

• infrastructure-standards

  Use when editing Proxmox/Terraform/Ansible inventory — VMID/IP assignment ranges and the Terraform-to-Ansible inventory contract.

• openbao-dynamic-aws-creds

  Replace a static AWS access key with short-lived STS creds minted by OpenBao's AWS engine via a credential_process wrapper. Covers architecture, bring-up, verification, upgrade caution. Use for a static key or credential_process diagnosis.

• openbao-secrets

  How to obtain and handle secrets under an OpenBao-backed model: pick the store tier, prefer engine-minted ephemeral creds over static, read pre-authorized, route write/apply via the human-gated secret_id. Use before fetching/wiring a credential.

• claude-skill-authoring

  Use when authoring or editing a skill in this estate — file layout, description length limits, and the self-contained-execution rule. General skill-authoring guidance is covered by Anthropic's own Agent Skills docs.

• nix-tool-policy

  Use when installing or choosing CLI tools in a Nix flake repo, editing flake.nix or home-manager config, or when tempted to pip/pipx/uv/brew/npm install anything. Tools come from the dev shell or nix shell — never ad-hoc package managers.

• skills-registry

  Use when looking up available tools, skills, commands, agents, or plugins

• workspace-standards

  Use when setting up or managing multi-repo workspaces

Similar plugins