← 返回
开发者工具 中文

Meta-Skill

Trajectory compiler that converts real OpenClaw session traces (or raw_trajectory_log) into a parameterized, reusable Skill via 4 stages: trace interception...
轨迹编译器,通过4阶段(轨迹拦截…)将真实OpenClaw会话轨迹(或原始轨迹日志)转换为参数化、可复用的Skill。
la1850 la1850 来源
开发者工具 clawhub v1.0.0 1 版本 99894.6 Key: 无需
★ 0
Stars
📥 948
下载
💾 8
安装
1
版本
#latest

概述

Meta‑Skill — Trajectory Compiler

Overview

Compile a concrete, parameter-filled execution trace into a reusable Skill with dynamic inputs. This skill implements a 4‑stage closed‑loop pipeline:

1) Trace Interception → 2) DAG Abstraction → 3) Schema+Code Synthesis → 4) Registration.

Stage 1 — Trace Interception (implemented)

Use scripts/trace-from-session.js to generate a real trace from OpenClaw session JSONL, or scripts/trace-interceptor.js for JSONL tool event streams.

Stage 2 — DAG Construction & Abstraction (implemented)

Use scripts/trajectory-compiler.js to build a DAG and lift variables to inputs.

Stage 3 — Code & Schema Synthesis (implemented)

Compiler emits references/schema.json, references/plan.json, references/run-flow.md, and scripts/run.js.

Stage 4 — Registration & Hot Reload (implemented)

Compiler writes into the Skills directory and requires a skills refresh.

Usage

See references/pipeline.md and references/compiler-spec.md.

版本历史

共 1 个版本

  • v1.0.0 当前
    2026-03-19 08:41 安全 安全

安全检测

腾讯云安全 (Keen)

安全,无风险
查看报告

腾讯云安全 (Sanbu)

安全,无风险
查看报告

🔗 相关推荐

ai-agent

self-improving agent

pskoett
捕获经验教训、错误及修正内容,以实现持续改进。适用于以下场景:(1)命令或操作意外失败;(2)用户纠正Claude(如“不,那不对……”“实际上……”);(3)用户请求的功能不存在;(4)外部API或工具出现故障;(5)Claude发现自身
★ 4,086 📥 814,738
ai-agent

Skill Vetter

spclaudehome
AI智能体技能安全预审工具。安装ClawdHub、GitHub等来源技能前,检查风险信号、权限范围及可疑模式。
★ 1,232 📥 268,303
ai-agent

Self-Improving + Proactive Agent

ivangdavila
自我反思+自我批评+自我学习+自组织记忆。智能体评估自身工作、发现错误并持续改进。
★ 1,385 📥 321,012