LightAgent: Lightweight Python framework for OpenAI-compatible agents with tools, memory, guardrails, tracing, lifecycle hooks, multi-agent collaboration, and workflows.
Our agent harness: Skills, plugins, hooks, and utilities to improve the quality of your agent.
Pre-compaction hooks + a CAS-guarded task pointer that keep Codex agents on-goal across context auto-compaction — no mainline drift, no duplicate work, no conflicting progress in long runs.
Reusable Agent Skills, plus integrations (statusline, provider usage, vision) that install into Codex, Claude Code, and opencode.
Stops AI coding agents from claiming work is done without evidence. Deterministic hooks decide, TypeSafe's Jev advises. Append-only ledger, zero runtime dependencies.
Make Opus 4.8 (or any Claude model) work like Claude Fable 5 — a Claude Code skill + guard hooks (plan gate, model ceiling, per-task enforcement) for Fable-5-grade discipline: plan, self-verify, and route sub-agents, without Fable 5.
Weft Coordination Protocol (WCP): live coordination for parallel coding agents across harnesses, built on Cloudflare. An Atelier (elier.ai) open-source project.
Waifu for hermes-agent. Dashboard plugin with multi-emotion support.
Route Codex PermissionRequest hooks to the BLE "buddy" hardware from anthropics/claude-desktop-buddy. Approve/deny Codex actions on a physical button. macOS, on-demand BLE.
YAML hook automation for the PI coding agent: tool guards, session hooks, prompts, notifications, and bash actions.
🍕 An opinionated, extensible Agent Harness based on Pi. Connect Multiple Computers to one harness and work anywhere.
Proof-carrying completion for AI coding agents. Block 'I'm done' until every critical obligation has evidence.
Multi-agent coordination + harness adapters + portable CLI utilities for Claude Code / Cursor / Codex.
Universal agent lifecycle / hook event handler for agent harnesses
Open protocol for agent-to-agent communication over HTTPS. No registry, no platform — just inboxes.
Evidence gates and session continuity for AI coding agents
Keep AI-generated UI on-system.
Be concise catches the agent's own verbosity and makes it corrects in real-time.
AI 抬头 | Mid-run self-review nudges for aimless AI coding agents: 检测重复命令/失败循环/零产出并注入提醒,真实本地时钟校准 AI 时间感 (ZCode / DeepSeek Harness plugin)
Wmux — Warp alert: native toasts when your AI agent (Claude, Codex, OpenCode, Gemini) finishes a turn.
To associate your repository with the agent-hooks topic, visit your repo's landing page and select "manage topics."