CANONICAL loctree agent plugin (suite). Structural perception — slice before edit, impact before delete, find before create (incl. multi-literal OR A|B on MCP/LSP/CLI), prism before vc-polarize. Powered by loctree-mcp + loctree-lsp. Deprecated peer path: suite ./plugin/ (redirect only).
Live crypto whale trades across 15 exchanges plus Ethereum/Base/Arbitrum DEX swaps, Smart Money Radar, liquidations from the venues that publish them, and market snapshots. Six analysis skills plus the hosted MCP server, no API key required.
★not rated 9▲
+1 27d agoA
tokens not measured
originalMIT
Self-improving robotics expertise for AI agents: skills for architecting, integrating, simulating, visualizing, testing, and training physical-AI applications (ROS 2, Nav2, Gazebo, LeRobot, Isaac, HuggingFace) — captured from real builds, mined from the ecosystem, evidence-verified.
★not rated 9 yesterdayA
tokens not measured
originalMIT
Several models on the same question at once — Claude sub-agents, OpenAI Codex and a cheap third via OpenCode. Review code (code-review), check whether the work is actually finished rather than finished-looking (check-if-done), brainstorm wide with a different model per cognitive frame (adhd), or just ask all of t.
★not rated 9▲
+1
changed todayA
tokens not measured
originalMIT
Installs graph engineering into a project that has none: turns an existing process — skills, a runbook, a CI config, or a description — into a guarded graph with typed nodes, total exit guards, bounded retry loops, a run-state file, a skipped-step ledger, its own auditor, and an executable eval suite. Domain-agnostic.
★not rated 9▲
+1 18d agoA
tokens not measured
originalMIT
The backend habits an AI assistant misses on its own — API versioning, feature flags, circuit breakers, and four more. Measured across 13 baseline runs; everything the model already does reliably was cut.
★not rated 9 12d agoA
tokens not measured
originalMIT
Voice coding for Claude Code, Jarvis-style — talk to your agent while it works. Ships the voice hooks and connects to the noisy-coding daemon (Docker or native).
★not rated 9▲
+2 yesterdayA
tokens not measured
originalMIT
Map animation skills for AI coding agents — build editorial, "Vox-style" animated maps for explainer and documentary video: camera zooms, drawn-on routes, dropped pins, region highlights, and the 12fps stutter.
★not rated 9▲
+2 6d agoA
tokens not measured
originalMIT
A session-based working method for coding agents: plan an epic into thin complete slices, build one at a time, verify by observing real behaviour, and hand off so the next session starts where the last one stopped.
★not rated 8 1mo agoA
tokens not measured
originalMIT
Audience-building workflows backed by the narrative-mcp server — turning marketing briefs into filtered, materialized audience datasets and activating them to downstream advertising connectors.
★not rated 8 yesterdayA
tokens not measured
originalMIT
A self-contained agentic development harness for Claude Code. One command runs the full feature pipeline: context, think, design + threat model, plan with self-review, sub-agent execution, double code review, and security gates. Bundles every skill and agent it needs.
★not rated 8 19d agoA
tokens not measured
originalMIT
Task-Track & Handoff for AI agents on long, multi-session work. Within a session it pins one ⭐️ MAIN goal (frozen title, explicit verdict at close) in an emoji-tagged task list so the agent doesn't drift; across sessions or agents it writes a plain-markdown handoff the next agent re-derives before trusting. Not for…
★not rated 8 19d agoA
tokens not measured
originalMIT
WAYD (What Are You Doing?). A deliberately minimal social feed for devs using AI coding agents: the scroll break you take while your agent works, without leaving your console. Share a vibe, react, reply, then get straight back to the task. Built on GitHub Issues as a transparent backend.
★not rated 8 3mo agoA
tokens not measured
originalMIT
Six X/Twitter skills for Claude Code — read tweets, search, bulk exports, real-time monitors, and posting through a connected account. Paid per call in USDC. No X developer account, no API keys, no OAuth.
rubber-duck. 🦆 Socratic assistant operating system for developers who want better-quality decisions, not blind automation.
★not rated 8 4d 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: