Power BI dashboard building for AI agents: the powerbi-agent MCP server (list workspaces and items, model schema, read-only DAX, download/validate/deploy PBIR reports with guardrails, date-filter pins, page captures, refresh history) and three skills (build, review, verify). Needs Python 3.11+, pip install -r requirem.
★not rated 1 16d agoA
tokens not measured
originalMIT
A coordination control plane for fleets of AI coding agents: claim without colliding, hand off bounded context, and finish only when the declared proof exists.
★not rated 1 4d agoA
tokens not measured
originalMIT
AI governance for agent tool calls: trust scoring, PII/harmful-content guardrails, hallucination detection, and compliance checks (NIST AI RMF, EU AI Act, ISO 42001) via 27 read-only MCP tools.
★not rated 1 yesterdayA
tokens not measured
originalMIT
Know Your Agent: fail-closed eval (ALLOW | DENY | REQUIREAPPROVE), dual-plane host=ide|runtime, human approval. This plugin is a thin MCP gate — the control plane is the sole PEP.
★not rated 1 yesterdayA
tokens not measured
originalMIT
412+ installable agent skills (SKILL.md) for professional apps, token efficiency, and common-sense prompting. Compatible with Gemini CLI skills discovery.
★not rated 1
changed todayA
tokens not measured
originalMIT
Self-contained Claude Code plugin that binds the canonical SameDayDesk MCP, one web-extract skill for single-page GET or bounded 1–5 URL batch POST, and one offline explicit-record skill. Unpaid discovery only. Mapping does not fetch or pay. No wallet, no second paywall, no installer.
★not rated 1
changed 5d agoA
tokens not measured
originalMIT
Give Claude a real browser: fetch bot-protected and JavaScript-heavy pages through UnblockingAPI, with structured JSON for any site a published template covers. Replaces the built-in web fetch.
★not rated 1
changed 3d agoA
tokens not measured
originalMIT
Connect Claude Code to 350+ business data sources via Windsor.ai. Query marketing, sales, CRM, ecommerce, finance, and analytics data from Google Ads, Meta, HubSpot, Salesforce, Shopify, Stripe, and hundreds more — directly from your terminal.
★not rated 0 1mo agoA
tokens not measured
originalMIT
Get named by AI search engines. Finds the questions where ChatGPT, Gemini and Perplexity name a competitor instead of you, writes the page that closes the gap, and measures what publishing it did.
★not rated 0 1mo agoA
tokens not measured
AGPL-3.0
Sandboxed AI-native bash for Claude Code: 70 MCP tools backed by the @ag-bash/bash engine — runbash with a full unix toolkit (grep/sed/awk/jq), surgical code edit/diff/analyze, WASM Python & JavaScript, semantic search, and agent task/worktree management. Defense-in-depth sandbox on by default.
★not rated 0 2d agoA
tokens not measured
originalApache-2.0
Strip AI tells from prose and put a human voice back, in British or American English, pitched at the reader it is for. A catalogue of tells across content, language, style, assistant artefacts, filler, condescension and overshare, four tone presets, three strength levels, a two-way politeness check that catches both…
Project-local Markdown progress tracking for resumable, multi-step agent work. Bundles the project-progress skill, an MCP server for a cross-project index, and lifecycle hooks that inject resume context on session start and guard sensitive progress on commit.
A bio-inspired context protocol for DataHub agents: budgeted foraging over the lineage graph, three-level semantic zoom, working memory that prunes itself, and reflex answers that never call a model. Extends the DataHub skills rather than replacing them.
★not rated 0 1mo agoA
tokens not measured
originalApache-2.0
Automatic long-term memory for Claude Code via Hindsight. Recalls relevant memories before each prompt and retains conversation transcripts after each response.
★not rated 0 17d agoA
tokens not measured
originalMIT
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: