🔌
Plugins
Loading...
Showing 30 of 22888 results
Generates UiPath Tasks workflows — Form Tasks, External Tasks, and Task Management activities for human-in-the-loop, system-in-the-loop, and recovery patterns. Plugs into uipath-core via plugin_loader.
Repo-agnostic pipeline agents, orchestrator, and a scouting initializer that generates per-repo skills + a routing inventory.
Transparent engineering orchestration with durable feature verification.
Maximum parallelism orchestration - spawns ALL independent tasks in a single message. Tool-native JSON outputs for certification.
Ultraprompt V9.4 cognitive control-plane plugin for Claude Code and Codex: 56 skills, 35 agents, 42 MCP tools, 33 commands, typed local memory, safe dream jobs, governed learning, workflow pathfinding, capability graph, live Mission Control dashboard, evidence-led release gates, dual-runtime install. V9.4 completes the release-integrity pass: one VERSION file feeds every version site; render-manifest --check fails CI on any version drift; the SubagentStart hook is single-sourced across POSIX and Windows; README and manifest catalog counts render from one template.
Deterministic markdown-based project memory: loads ai/*.md context at session start and provides a skill for keeping it up to date. No AI/LLM calls.
Deploy, configure, and manage Umami — open-source privacy-focused web analytics with API client, tracker, events, statistics, reports, and team management
Umbraco backoffice skills covering Foundation, Extension Types, Property Editors, Rich Text, Backend, and Setup
Umbraco backoffice testing skills covering unit, MSW, mocked repo, Mocked backoffice and e2e testing using playwright
Expert assistant for developing, packaging, testing, and submitting apps for umbrelOS
Deflates LinkedIn and corporate thought-leadership posts back to what actually happened: a plain rewrite, a brutal one-liner, a 💩 bullshit meter, and the manipulation tells — each grounded in real persuasion research.
Complete-output discipline — no truncation, no placeholders, no "for brevity". Fully standalone; the sutra orchestrator applies it to full-output steps.
Automate Unanet timesheet entry via Playwright and Okta SSO with 1Password credentials
from unclecode/claude-code-tools
Pointer to Understand-Anything (Egonex-AI, MIT) — a multi-agent pipeline that turns a codebase into an interactive knowledge graph with guided tours, semantic search, diff-impact analysis, and an onboarding dashboard. Its analysis core is a TypeScript + native tree-sitter build, so it is installed upstream rather than re-homed into this markdown marketplace (same posture as not re-homing claude-mem's runtime). This entry documents what it does, when to reach for it vs. graphify / github-repo-analyzer, and the one-line install. A full native vendor is a follow-up decision (DR-011).
AI-powered codebase analysis that produces interactive knowledge graphs, guided tours, and plain-English architecture explanations.
Skills that turn dense code and concepts into structured, navigable explanations — interactive box-diagrams of code and semantic structure-aware diffs. For when you want to understand, not just read.
Improve LLM apps and agents from real traces — cost/latency/quality, evals, GEPA, model comparison, and routing.
Write code, comments, tests, or commit messages that read as if the project's own maintainer wrote them, so a contribution is accepted as native rather than rejected as foreign. A foreign-looking patch is a cost to the maintainer even when it's correct.
On-device audio transcripts you own. Ingest local files, YouTube, podcasts, RSS, Quill, or Google Meet, run local ASR, diarization, and speaker fingerprinting, then browse, search, and export. One skill that routes to a reference per surface.
Guided reflective journaling with evidence-based psychological frameworks and optional clinical screening
Crawl university admission/program data via natural language. Skills cover install, crawl, diagnose, and export.
A harness that makes Opus (or any Claude/Codex model) behave like Fable: completion, evidence, and verification enforced as procedure, auto-routed per task (render-output verification, multi-story evidence gate, investigation protocol, early-stop guard). Runs on both Claude Code (plugin) and Codex (skill + hooks). Fork of fivetaku/fablize, fixing the PostToolUse false-positive failure gate.
UniFi Network and Protect CLI for managing UDM devices, clients, VLANs, firewall rules, traffic routes, cameras, PTZ control, and motion events — dry-run by default, API key auth, no MCP supply chain risk
Write idiomatic unihttp client code and generate typed unihttp client packages (SDKs) from an OpenAPI spec or an API description