MCP server for EVM security analysis, gas profiling, compiler diagnostics, and transaction simulation. Provides autonomous coding agents with deterministic, schema-validated tools for Ethereum smart contract development.
★not rated 1 2mo agoA
tokens not measured
originalMIT
Encoding-transparent file tools for legacy codepages in Claude Code. Adds Read, Grep, Edit, and Write over MCP with .encoding-rules-aware decoding and write-back.
★not rated 1 14d agoA
tokens not measured
originalMIT
Turn your AI Agent into a Power BI analyst. 4 skills (kpim-analysis · powerbi-pipeline · powerbi-mcp · powerbi-knowledge), 8 /powerbi- commands, and the powerbi-knowledge-curator agent. For the 16 MCP tools (DAX/model/report/knowledge) run install.ps1 in the repo root — this plugin ships the workflows & knowledge laye.
Coordinate and supervise parent/child sessions across Claude, Codex, OpenCode, ZCode, Fast Agent, Hermes, and Qoder with MCP messaging, TTL coordination notes, session inspection, and autopilot.
★not rated 1
changed 18d agoA
tokens not measured
originalMIT
Let one agent dispatch the agy CLI (Antigravity, Gemini-backed): ask a model, continue/resume conversations, list models, check auth/credit, log in, and generate images via agy's native model.
★not rated 1 2mo agoA
tokens not measured
originalMIT
Co-edit a live tldraw whiteboard with Claude — boxes, UML classes, sticky notes, arrows, and server-side auto-layout. Boots a local sync backend + web UI (http://127.0.0.1:5858) so you watch Claude draw in real time in your browser.
Manage Google Cloud TPU capacity and Gemma 4 vLLM serving on TPU VMs: provisioning, queued resources, quotas, cost estimates, benchmarking, and SRE diagnostics via the tpu-devops MCP server.
Offload repository exploration, semantic search, code-graph queries, test generation, docs, and lint or type fixes to a model you run yourself on LM Studio, Ollama, vLLM, or LocalAI. Returns validated, unapplied diffs; repository content never leaves your network.
★not rated 1 1mo agoA
tokens not measured
originalMIT
Pick a topic and the agent researches sources, scores them, and builds a 30-day course you study one day at a time. Runs entirely on your own machine; the skill starts the local server itself.
★not rated 1 1mo agoA
tokens not measured
originalMIT