skill-extras
claudeUpstream skill compositions: adversarial grilling harness (/grill-harness), triage disclaimer suppression (/triage-override), code-review disclaimer suppression (/code-review-override).
3 skills
Git workflow automation: gentle branch cleanup (/tidy), unified MR creation (/ship), session statistics (/stats), issue spec-to-MR pipeline (/issue-to-spec, /spec-to-mr), review-feedback follow-through (/mr-followthrough), and peer-review feeding with post-back (/peer-review-feed).
Plugin dev-workflow from marketplace vvaldez-agent-skills in vvaldez/agent-skills. Commands read from the repo's manifest — yesterday.
drainDrain the current session's residue into its durable sinks — audits what is NOT yet captured and routes each item: new lessons → propose to LESSONS.md, undrained gaps → proposed GitLab issues, git state (merged/stale worktrees and branches, stray files) → reported with recommended actions. Always en
issue-to-specTake an underspecified GitLab issue from triage to a fully spec'd, agent-ready issue in one invocation. /issue-to-spec <issue-url> [more-urls...] — reads the issue, its references, and all comments, grills the user interactively to close every spec gap (problem, deliverables, acceptance criteria, sc
mr-followthroughDrive a GitLab MR from reviewer feedback to merged in one invocation. /mr-followthrough <mr-url> [more-urls...] [--merge] — reads the reviewer's TRACK/NOTE findings, triages dispositions with the user, implements fixes, re-reviews only the delta (single, dual if security-sensitive), pushes, waits fo
peer-review-feedFeed a peer-mode dual review + findings post-back to the configured agent agency in one invocation. /peer-review-feed <mr-url> [--disposition] [--agent <name>] — resolves the MR, refuses self-authored MRs (findings stay in-session), creates the agency card with the GENERATED posting contract (one st
shipUnified MR/PR workflow — detects git state, creates a branch if on main, commits with conventional format, pushes, and creates a GitLab merge request (glab) or GitHub pull request (gh) in one command. Auto-detects platform from remote URL. Use when user says /ship, "create MR", "create PR", "ship th
spec-to-mrTake a spec'd GitLab issue from ready-for-agent to pushed MR in one invocation. The second half of the issue pipeline: /issue-to-spec defines the spec (grills the user, writes it into the issue description, labels workflow::ready-for-agent), then /spec-to-mr <issue-url> [more-urls...] reads the spec
spec-to-readyTake a workflow::ready-for-agent GitLab issue to an operator-ready MR through a three-stage relay of explicitly different agents. /spec-to-ready <issue-url> — stage 1 runs /spec-to-mr (implement, review, verify, push the MR, ship-and-stop); stage 2 feeds a code-review TRIPLE ULTRA (secops + devops +
statsSession statistics dashboard — parses Claude Code JSONL transcripts or the OpenCode session store (auto-detected) to show slash command usage, tool call counts, MRs/PRs created, commits, agent spawns, doc edits, and session duration. Use when user says /stats, "session stats", "what did I do", "how
tidyGentle git housekeeping — fetches origin, prunes stale remote refs, reports branch status, auto-deletes merged branches, and confirms before touching unmerged ones. Non-destructive by default. Use when user says /tidy, "clean up git", "check branches", "housekeeping", "are my branches clean", or aft
Upstream skill compositions: adversarial grilling harness (/grill-harness), triage disclaimer suppression (/triage-override), code-review disclaimer suppression (/code-review-override).
3 skills
Ansible automation skills: audit and refactor codebases to follow battle-tested Ansible patterns (/ansibleize).
1 skill
Red Hat branded process flow diagrams as self-contained HTML with auto-PNG export (/diagram).
1 skill
Session handoffs as GitHub issues: extends /handoff to publish a labeled, close-when-consumed issue instead of a markdown file, with dedup against existing tracked work (/handoff-issue).
1 skill
Offload bulk, low-judgement text work to a locally-hosted model so it never enters the agent's context: cheap first-pass diff review (/local-review). NOT SELF-CONTAINED - requires the separate localllm tooling, which is not bundled here and not publicly available, plus a running local model worker, Windows and PowerShell 5.1. Without those the skill reports that it cannot run and stops.
1 skill
Deck publishing tools: convert a markdown deck (quick-deck companion markdown or generic markdown) into a native, co-editable Google Slides presentation via a deterministic pandoc + Drive-conversion pipeline, with built-in post-upload verification (/deck-to-gslides).
1 skill
Talk like caveman. Cut filler. Keep technical details explicit.
22 skills · 1 command · 3 agents · hooks
Persistent memory for Claude Code: the next session starts where the last one left off.
29 skills · 1 MCP server · hooks
tasteskill 2.5.1: professional company websites at the figma.com and x.ai level, with one section per screen, real photographs, real product UI, effects kept for tech and gaming pages, and a real-browser checker and finish gate.
1 skill
Open-source AI job search that runs locally in your AI coding CLI: scan job portals, evaluate listings into a structured A-H report with a 1-5 score, tailor your CV, track applications
5 skills · hooks
Cross-session memory and token savings for coding agents.
6 skills · 1 agent · 1 MCP server · hooks
Package any website or local web build into a lightweight desktop app with Pake CLI.
1 skill
This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.