🔌
Plugins
Loading...
Showing 30 of 23013 results
Build a quick repo tour with architecture, entry points, and likely edit paths
Preferred GitHub repository standards, audited/applied with `gh` — a repo-settings baseline (wiki/projects off, auto-delete merged branches, read-only workflow token, SHA-pinned actions, Dependabot security updates), three branch-ruleset classes, a Dependabot version-updates config, a lightweight tier for content/asset/config repos, and the README/CONTRIBUTING/CLAUDE.md doc split. Ships templates plus a read-only repo-standards-auditor agent for per-repo (fan-out) conformance audits.
審查並套用 JurisLM repo 標準 — release workflow、ESLint、worktree、Bun runtime、Vitest
Maintain a Claude-written summaries sidecar in any repo — a committed read-side module plus a config-driven daily refresh skill that summarizes real commit and release data into one-line suffixes.
LLM wiki for code repos — init from src + commits, grow from changes, query in natural language.
Scaffold standard-compliant repos (/new-repo) and apply the repo standard while you work.
Adds RepoPrompt guidance at session start and after compaction only when rp-cli is reachable.
AI agent that automates distribution of consolidated sales reports to representatives based on territorial parameters
Local-first 可稽核技術報告編譯器:宣告式 DOCX、需求證據追溯、人工審查、政策包、QA 與 release bundle。
Report Me — turn a topic or your materials into a single self-contained, interactive HTML report (tabs · reference drawers · heatmaps · SVG · in-page search) in your language, verified by a multi-agent QA harness. Opens from disk, no deployment. Invoke with /reportme.
Workflow skills for the Reprise MCP v2 (atomic per-action surface).
Prompt engineering skill for Claude Code, OpenClaw, Codex, Grok CLI, and Hermes Agent
RepuSw - Enhanced development workflow plugin with code review, status checks, and more
Interactive development-requirements capture. /req [what] interviews the requester and writes a requirements document; /req-brainstorm [request] runs active-proposal brainstorming before producing the same artifact; /req-lite [request] auto-drafts a requirements doc for lightweight maintenance requests; /req-setup initializes a per-product context workspace at docs/req-skill/.
Pipeline iterativo de ingenieria de requisitos (LEL y Escenarios): descubre el mapa del producto desde documentos, carpetas o una entrevista sin documento; elabora y baselinea features por incrementos; y absorbe cambios con confirmacion y changelog, todo trazable.
Deep-research harness: plan conversationally, execute autonomously into a cited report with resumable file state and pluggable backends.
Evidence-grounded research portable across every repo. Eleven skills (scoping, source curation, synthesis, adversarial review, decision support, rationale reconstruction, plus a four-skill project-mode lifecycle for sustained multi-week investigations) plus two retrieval subagents; methodology grounded in seven convergent disciplines (STORM, PRISMA, ACH, Wikipedia V/RS/NPOV, OSINT, GIJN, GRADE).
Project-scoped knowledge base for Claude Code with progressive disclosure (INDEX.md dispatcher, sed-extracted summaries) and contrarian-pass investigation via background Opus subagents. Conflict-handling history prevents re-trying refuted approaches across sessions.
PhD-level multi-perspective research inspired by Stanford's STORM. The 'storm research' skill orchestrates five expert persona agents (practitioner, skeptic, economist, historian, academic), maps their contradictions, synthesizes a cited briefing, and red-teams the result. Say 'storm research <topic>' to start.
Discovery-to-watchlist research primitives. idea-search is a read-only scan over the agentm recall engine surfacing existing vault/codebase entries relevant to a question. learn-forward leans by name on agentm's approved-source forward-learning pipeline to mine operator-configured sources onto the watchlist. codebase-improvement applies a research insight's stale pattern against the operator's own repo, surfacing exactly one watchlist finding per match -- never auto-editing. All three are strictly discovery-surfacing: findings land in the watchlist for operator review, nothing is ever auto-adopted.
Investigate a question against high-trust primary sources and capture the findings as a Markdown file in the repo.