{"data":[{"slug":"total-recall","title":"total-recall","description":"Coding Agents forget everything when a session ends. total-recall fixes that: a shared memory layer that persists across sessions, tools, and devices.\n\nMemory is organized into three tiers (Hot / Warm / Cold) with a 4,000-token budget enforced automatically — so Claude gets relevant context injected at session start without carrying stale noise. Semantic search (all-MiniLM-L6-v2 embeddings, offline) retrieves the right memories on demand.\n\nA knowledge base lets you ingest READMEs, API docs, and architecture notes that are retrieved semantically when relevant. An eval framework lets you measure retrieval quality, run benchmarks, and compare config changes — so you know memory is actually helping.\n\nAll state is local: SQLite + ONNX embeddings, no external services, no API keys required.\n\nWorks across Claude Code, Copilot CLI, Cursor, Cline, and OpenCode — existing memories auto-import on first run.","developer":"strvmarv","category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-11-12T00:07:47.335Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/strvmarv/total-recall.git"},{"slug":"superpipelines","title":"superpipelines","description":"Multi-agent AI pipelines with structural write/review isolation, crash-safe state recovery, and spec-driven development. Run sequential, parallel fan-out, iterative-loop, or human-gated patterns across Claude Code, Cursor, Codex, OpenCode, Copilot CLI, and Gemini CLI.","developer":"Gustavo Meilus","category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-11-12T00:07:47.335Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/gustavo-meilus/superpipelines.git"},{"slug":"skillers","title":"skillers","description":"Learn from your workflow patterns and suggest skills, hooks, and agents to automate the repetitive work. Skillers reads your AI conversation transcripts (Claude Code, Codex CLI, OpenCode), extracts recurring patterns into themed weighted-knowledge files, then recommends concrete skills/hooks/agents to codify them. Two phases: `/skillers compact` analyzes transcripts and writes `.skillers/<theme>.md` knowledge files; `/skillers recommend` reads accumulated knowledge plus your existing plugin/skill ecosystem and suggests automations that don't already exist. The output is \"here are 4 patterns I see you doing manually every week and the exact SKILL.md / hooks.json that would automate each.\"","developer":"Avi Fenesh","category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-11-12T00:07:47.335Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/agent-sh/skillers.git"},{"slug":"phone-a-friend-paf","title":"phone-a-friend-paf","description":"Phone a Friend lets Claude Code call other AI coding backends for second opinions, code review, and multi-agent collaboration. It relays your prompt and repository context to Codex, Gemini, OpenCode, or local Ollama models, then brings the response back into your Claude session. Includes streaming, structured JSON output, session continuity, multi-agent orchestration with a web dashboard, and an interactive TUI. Useful when you want a different model's perspective without leaving Claude Code.","developer":"Bruno Freiberger","category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-11-12T00:07:47.335Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/freibergergarcia/phone-a-friend.git"},{"slug":"phone-a-friend","title":"phone-a-friend","description":"Phone a Friend is a CLI relay that lets AI coding agents collaborate by sending prompts and repository context to other AI backends. It supports Claude, Codex, Gemini, and Ollama as relay targets. Agents can get second opinions, run multi-agent sessions with persistent communication, and stream responses in real time. It includes an interactive TUI dashboard, TOML configuration, and agentic orchestration mode for multi-agent collaboration.","developer":"Bruno Freiberger","category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-11-12T00:07:47.335Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/freibergergarcia/phone-a-friend.git"},{"slug":"opinionated-prisma-skills","title":"opinionated-prisma-skills","description":"Opinionated skill plugin for Prisma ORM on self-hosted PostgreSQL. Enforces consistent decisions across schema design, migration safety, indexing, transactions, and the raw SQL boundary. Covers UUID v7, snake_case mapping, foreign key indexes, three-step safe migrations, concurrent index creation, and transaction patterns. Skills activate automatically based on context. Supports Claude Code, Codex CLI, Gemini CLI, Cursor, and OpenCode.","developer":"Morley Media Group","category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-11-12T00:07:47.335Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/esot321c/opinionated-prisma.git"},{"slug":"nexus-agents","title":"nexus-agents","description":"Intelligent orchestration platform for AI coding tools. Coordinates Claude, Codex, Gemini, and OpenCode through a single MCP server — routing each task to the best model using data-driven algorithms (LinUCB bandit, TOPSIS scoring), validating outputs through multi-model consensus voting, and continuously improving through outcome-driven learning. Features 29 MCP tools, 11 expert types created on demand, 8 memory backends with cross-session persistence, a full dev pipeline (research, plan, vote, implement, QA, security), and research discovery from 9 sources including arXiv and GitHub.","developer":"William Zujkowski","category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-11-12T00:07:47.335Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/williamzujkowski/nexus-agents.git"},{"slug":"learn","title":"learn","description":"Research any topic online and create comprehensive learning guides with RAG-optimized indexes for your AI agents. `/learn react hooks` searches the web for ~20 sources, scores each on a 100-point scale across authority (3x), recency (2x), depth (2x), examples (2x), and uniqueness (1x), fetches only high-scoring sources, and writes a synthesized guide to `agent-knowledge/<topic>.md` with companion `resources/<topic>-sources.json` containing full source metadata. Six-stage methodology: progressive discovery, quality scoring, just-in-time extraction, synthesis, source citation, RAG optimization. Output shape designed for ingestion by Claude Code, OpenCode, and Codex agent-knowledge libraries.","developer":"Avi Fenesh","category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-11-12T00:07:47.335Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/agent-sh/learn.git"},{"slug":"lean-kg","title":"lean-kg","description":"Lightweight Knowledge Graph for AI-Assisted Development\nLeanKG is a local-first knowledge graph that gives AI coding tools accurate codebase context. It indexes your code, builds dependency graphs, and exposes an MCP server so tools like Cursor, OpenCode, and Claude Code can query the knowledge graph directly. No cloud services, no external databases.\nVisualize your knowledge graph with force-directed layout, WebGL rendering, and community clustering.","developer":"FreePeak","category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-11-12T00:07:47.335Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/FreePeak/LeanKG.git"},{"slug":"docs-context","title":"docs-context","description":"docs-context is the \"Super Base Context\" for Agent Coding: a structured replacement for single-file CLAUDE.md that separates project knowledge into five purpose-specific docs (architecture, modules, coding, tech-stack, decisions), loads only what the current task needs, and keeps those docs in sync with the code as a net-delta-triggered Write mode.\n\nRead mode fires before any non-trivial write/modify/refactor/design/plan task, grounding the agent in architectural constraints and conventions. Write mode fires once at the end of a task when landed changes (code, config, SQL, build scripts) differ from the doc state — no firing on read-only work, in-progress edits, or try-and-undo.\n\nWorks across Claude Code, Codex CLI, Cursor, OpenCode, Antigravity, Trae, Qoder via the standard SKILL.md format.","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-11-12T00:07:47.335Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/lujiafa/houtu-project-skills.git"},{"slug":"debate","title":"debate","description":"Structured multi-round debate between AI tools with proposer / challenger roles and a final verdict. `/debate` orchestrates a back-and-forth between two AI CLIs (Gemini, Codex, Claude Code, OpenCode, Copilot, Kiro) on a design decision, technical question, or contested approach. Each round, one tool proposes / defends and the other challenges; the orchestrator synthesizes the disagreement and renders a verdict with reasoning. Surfaces blind spots that single-tool consultation misses, especially on architectural decisions where two strong arguments exist. Programmatic entry point — other agents and workflows can spawn a debate via Task() to break a tie.","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-11-12T00:07:47.335Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/agent-sh/debate.git"},{"slug":"ai-sync","title":"ai-sync","description":"Cross-platform AI agent synchronization. When you switch \n  between AI coding tools (Claude Code, Cursor, Codex,     \n  Aider, Cline, Windsurf, Copilot), ai-sync maintains a    \n  shared .ai-sync/ directory so the next agent picks up    \n  exactly where the last one left off — with full context  \n  of what was done, what's next, blockers, and build       \n  status. Features conflict detection, auto-progress       \n  tracking, session audit trails, and adapter files for 9  \n  tools.","developer":"Oreolion","category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-11-12T00:07:47.335Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/oreolion/ai-sync-plugin.git"},{"slug":"ai-dev-kit","title":"ai-dev-kit","description":"AI Dev Kit is a complete plugin-oriented developer workspace for AI coding assistants. It ships everything needed to run professional-grade engineering workflows — TDD, code review, security audit, CI/CD, ML pipelines, infrastructure-as-code, code quality analysis, and debt tracking  across Claude Code, Codex, OpenCode, and Gemini CLI.\n\nThink of it as a system prompt that scales to 59 specialized skills and 33 domain agents, with lifecycle hooks and automated validation.","developer":"Noah Sheldon","category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-11-12T00:07:47.335Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/noah-sheldon/ai-dev-kit.git"},{"slug":"ai-coding-ok","title":"ai-coding-ok","description":"PDCA memory loop for Claude Code — prevents \"AI fixed bug X and broke feature Y\" across iterations.\n\nInstalls a three-tier memory system in your project (project-memory, decisions-log, task-history) and enforces a closed PDCA loop on every task: Plan (read memory before coding), Do, Check, Act (write memory back after). The Act step is what most tools skip — without it, your context file rots after 10 iterations. With it, iteration 50 has the same context quality as iteration 1.\n\nComplements superpowers: superpowers brings per-session discipline, ai-coding-ok brings cross-session memory. Also works with Copilot, Cursor, and OpenCode via the same templates.","developer":"Mark7766","category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-11-12T00:07:47.335Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/Mark7766/ai-coding-ok.git"},{"slug":"agnix","title":"agnix","description":"Lint agent configurations before they silently fail. 414 validation rules across Claude Code, Codex CLI, OpenCode, Cursor, GitHub Copilot, Gemini CLI, Cline, Windsurf, Roo Code, Amp, and Kiro - validates CLAUDE.md, AGENTS.md, SKILL.md, hooks.json, MCP server configs, and plugin manifests. Auto-fix included for safe issues; SARIF output for GitHub Code Scanning. Distributed as CLI (npm + Cargo), official GitHub Action (`agent-sh/agnix`), and IDE extensions for VS Code, JetBrains, Neovim, and Zed. Rules sourced from official specs, academic research on agent prompt design, and real-world failure patterns from production deployments. Mentioned in awesome-claude-code.","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-11-12T00:07:47.335Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/agent-sh/agnix.git"},{"slug":"agentmancy","title":"agentmancy","description":"AI Agent & Skill Builder for Claude Code, OpenCode, and Gemini.\n\nA free alternative to summonaikit that leverages your already existing subscriptions to Claude.","developer":"vinngo","category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-11-12T00:07:47.335Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/vinngo/agentmancy.git"},{"slug":"agent-knowledge","title":"agent-knowledge","description":"Cross-session memory for AI coding agents. A git-synced Markdown knowledge base with hybrid semantic + TF-IDF search, typed knowledge graph (11 edge types with directed BFS traversal), confidence/decay scoring, and automatic session distillation with secrets scrubbing. Reads session history from Claude Code, Cursor, and OpenCode.","developer":"keshrath","category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-11-12T00:07:47.335Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/keshrath/agent-knowledge.git"}],"total":17,"page":1,"per_page":36,"observed_at":"2026-11-12T00:07:47.335Z","source_counts":{"web":529,"demo":12,"skills":5,"official":317,"community":2283,"knowledge-work":123}}