Continuous Evolution

Automatically records experience, analyzes skill gaps, updates capability scores, and generates priority tasks to ensure continuous improvement after each task.

Security Scan

The skill's purpose (continuous improvement) matches the included script's behavior, but the runtime artifact (a shell script) writes into a root workspace, assumes external binaries, and the SKILL.md is high-level and does not explicitly document running that script — these mismatches warrant caution.

Purpose & Capability

Name/description (continuous improvement after each task) is consistent with the included evolve.sh which records experience, analyzes gaps, updates assessment, and emits P1 evolution tasks. The files SKILL.md describes (experience/gaps/assessment and P1 task JSON) match what the script produces.

Instruction Scope

SKILL.md contains high-level runtime instructions and expected output files but does not explicitly instruct executing evolve.sh. The included evolve.sh, if executed, writes files into absolute paths under /root/.openclaw/workspace (tasks, memory, skills). Writing into an agent root workspace and touching a 'skills' directory increases the surface for accidental access to other skill state; the script also generates P1 tasks automatically which could influence agent task scheduling. The SKILL.md and the runnable script are not fully aligned in execution detail, leaving agents discretion that could be broad.

Install Mechanism

No install spec (instruction-only at registry level) and no network downloads — low install risk. The repo includes a local shell script (evolve.sh) which will be executed from disk if the agent chooses to run it; no external installers or remote fetches are used.

Credentials

The skill asks for no environment variables or credentials, which is appropriate for its stated purpose. However, evolve.sh assumes availability of utilities (jq, bc, awk) without declaring them; it also writes to /root/.openclaw/workspace which may contain other agents' or skills' data — this is more privileged than writing to a per-skill sandboxed directory and should be validated.

Persistence & Privilege

always: false (good). Nevertheless, the script writes persistent files into a root workspace and generates high-priority tasks (P1) in tasks/queue; combined with autonomous execution this could allow the skill to keep creating and enqueuing tasks that affect the agent's workload. The skill does not modify other skills' configs directly, but the path choices allow proximity to other skill artifacts.

What to consider before installing

This skill is likely doing what it claims, but exercise caution before enabling it automatically. Before running: (1) Inspect evolve.sh yourself — it creates files under /root/.openclaw/workspace (tasks, memory, skills) and will generate P1 tasks; ensure those paths are acceptable. (2) Run the script in a non-production or sandboxed environment first to confirm behavior. (3) Ensure necessary binaries (jq, bc, awk) exist or edit the script to avoid unmet dependencies. (4) If you want least privilege, change TASKS_DIR/MEMORY_DIR to a per-skill sandbox (not /root) so it cannot access or confuse other skills' data. (5) Consider requiring explicit user invocation (not autonomous) until you trust its behavior and avoid letting it auto-enqueue high-priority tasks.

🔄 持续进化系统

核心理念

每一次任务执行，都是进化的机会！

---

🎯 进化机制

1. 经验记录 (每次任务后)

任务完成
    ↓
记录执行经验
    ↓
分析成功/失败原因
    ↓
归档到经验库

2. 能力缺口分析 (自动)

检查任务类型
    ↓
识别能力不足
    ↓
记录到缺口清单
    ↓
生成进化任务

3. 能力评估更新 (实时)

收集审计评分
    ↓
计算平均分
    ↓
更新能力等级
    ↓
可视化展示

4. 自动进化任务 (低分触发)

审计评分 < 70
    ↓
自动生成进化任务
    ↓
加入任务队列
    ↓
优先执行

---

📊 能力评估模型

评分等级

平均分	等级	说明
≥80	Advanced	高级 AI 工程师
60-79	Intermediate	中级 AI 工程师
<60	Basic	初级 AI 工程师

评估维度

任务理解能力 (权重 20%)
市场调研能力 (权重 15%)
自主学习能 力 (权重 15%)
执行能力 (权重 20%)
代码能力 (权重 10%)
数据分析能力 (权重 10%)
报告生成能力 (权重 10%)

---

📁 输出文件

文件	说明
memory/evolution/experience-YYYY-MM-DD.md	每日经验记录
memory/evolution/gaps-YYYY-MM-DD.md	能力缺口清单
memory/evolution/capability-assessment.json	能力评估
tasks/queue/P1_evol-*.json	进化任务

---

🔄 工作流程

任务执行完成
    ↓
质量审计 (评分)
    ↓
持续进化模块
├─ 记录经验 ✅
├─ 分析缺口 ✅
├─ 更新评估 ✅
└─ 生成进化任务 (如需要) ✅
    ↓
进化任务加入队列
    ↓
下一轮迭代执行
    ↓
能力持续提升

---

📈 进化追踪

经验积累

• 每日经验记录
• 成功案例归档
• 失败教训总结

能力提升

• 实时能力评估
• 能力曲线可视化
• 等级晋升追踪

进化任务

• 自动生成
• 优先级 P1
• 针对性增强

---

🫡 总裁，这样就能：

1. 每次任务都进化 - 不是阶段性，是持续的
2. 自动发现不足 - 低分自动触发进化任务
3. 能力可视化 - 实时看到能力提升
4. 经验不丢失 - 所有经验都记录归档

---

持续进化系统已集成！ 🫡