byted-ark-evolve
Agent self-evolution system.
Skill by volcengine/modelarkOfficial
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/bytedance-byted-ark-evolve
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: frontend, agent-building
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
- Runs as
- Python package
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/byted-ark-evolve
What's inside
- Skill name
byted-ark-evolve- Description
- Agent self-evolution system. Collects signals from user feedback, stores execution trajectories (golden + correction pairs), analyzes patterns, proposes mutations to workspace files, and generates HTML reports. Trigger with "/evolve" or when the user asks to improve agent behavior. Also use when the user says "remember this pattern", "don't do X again", or "that was a good approach, save it".
- License
- Apache-2.0
- Metadata
- tags: evolution meta-skill self-evolving · author: volcengine/modelark · version: 0.3.1
- Instructions
- 473 lines in SKILL.md · view on GitHub
Before installing
- Ships executable scripts
- Instructions call the network
- Instructions read files in your home folder (~/.ssh, ~/.aws…)
Files (41)
- LICENSE
- README.md
- SKILL.md
- dashboard-schema-example.json · references/
- evolution-steps.md · references/
- file-semantic-map.md · references/
- gene-contract.md · references/
- gene-library.json · references/
- init-rules.md · references/
- layer-model.md · references/
- onboarding.md · references/
- pareto-rules.md · references/
- pending-proposal-lifecycle.md · references/
- report-schema-example.json · references/
- signal-types.md · references/
- trajectory-templates.md · references/
- _workspace.py · scripts/
- apply-proposal.py · scripts/
- dashboard-render.py · scripts/
- dashboard-template.html · scripts/
- db-init.py · scripts/
- db-migrate-source.py · scripts/
- db-migrate-v031.py · scripts/
- detect-signal.py · scripts/
- gate-check.py · scripts/
- gene.py · scripts/
- orchestrator.py · scripts/
- pending-evolution.py · scripts/
- query.py · scripts/
- report-gen.py · scripts/
- report-render.py · scripts/
- scan-history.py · scripts/
- signal-record.py · scripts/
- snapshot.py · scripts/
- track-change.py · scripts/
- trajectory-skill-check.py · scripts/
- workspace-init.py · scripts/
- README.md · test-conversations/
- conversation-1-research-before-asking.md · test-conversations/
- conversation-2-memory-access-boundary.md · test-conversations/
- conversation-3-report-usability.md · test-conversations/
SKILL.md (first lines)
# Evolution Skill
Agent 自进化系统。进化单位是整个 Agent(Identity + Context + Protocol + Capability + Runtime)。
## 核心原则
1. **Quality > Reliability > Efficiency > Cost** — 不允许牺牲质量换效率
2. **帕累托约束** — 任何 mutation 如果导致某个维度退化,BLOCK 并请用户决策
3. **写入即生效** — 用户确认变更后写入 workspace 文件,下次 session 即可加载新配置
4. **先对齐再执行** — 进化方案必须先呈现给用户确认
## 数据安全与权限边界
- **本地存储**:所有运行时数据(信号、轨迹、变异、报告)写入用户本地 `~/.{*claw*}/workspace/evolution-data/`,不上传任何外部服务。
- **无外网调用**:脚本不发起 HTTP/socket 请求(可代码层验证:无 `requests` / `urllib` / `http.client` / `socket` 等导入用于网络通信)。
- **变更前置确认**:mutation 写入 workspace 文件前,必须由用户在对话中显式接受提案;用户未接受前不会修改任何文件。
- **作用范围**:写入仅限 `~/.{*claw*}/workspace/` 目录及其子目录;不访问 `/etc/`、`~/.ssh/`、`~/.aws/` 等敏感路径。
- **可审计**:每次 apply 通过 git commit 留痕,用户可随时 `git log` 查看完整变更历史。
- **可追溯**:每条 mutation 在 SQLite DB 中保留 source(evolution / user-direct / snapshot-diff)、决策状态、时间戳。
- **可禁用**:删除 `evolution-data/` 目录或卸载 skill 即完全停止;删除 `~/.{*claw*}/workspace/` 不会影响其他 skill。
## 数据目录
> **Runtime 适配**:workspace 根路径自动探测 `~/.{*claw*}/workspace/`(优先 `.arkclaw` → `.openclaw` → 其他含 `claw` 的目录),也可用 `CLAW_WORKSPACE` 环境变量显式指定。新装默认创建在 `~/.arkclaw/workspace/`。
```
evolution-data/
├── file-registry.json ← 初始化扫描结果
├── evolution.db ← SQLite(信号、变异、轨迹)
├── snapshot.json ← Workspace 文件 hash 快照(兜底变更检测)
├── trajectories/
│ ├── golden/ ← 正确执行轨迹
│ └── corrections/ ← 错误→修正对
├── dashboard.html ← Dashboard 页面(dashboard-render.py 自动生成)
└── reports/
├── evolution-*.json ← 进化报告数据(JSON)
└── evolution-*.html ← 进化报告页面(渲染后)
```
## References(按需读取,不随 session 加载)
| 文件 | 内容 | 何时读取 |
|------|------|---------|
| `references/layer-model.md` | 5 层模型 + 归因规则 | 进化分析时 |
| `references/pareto-rules.md` | 帕累托约束 + 验证标准 | 进化分析时 |
| `references/file-semantic-map.md` | 已知文件→层/语义/风险映射 | 初始化 + 归因时 |
| `references/init-rules.md` | 状态判定规则 + 占位符列表 | 初始化时 |
| `references/signal-types.md` | 信号识别规则 + 示例 | 信号收集时 |
| `references/trajectory-templates.md` | 轨迹存储格式 + 示例 | 轨迹写入时 |
| `references/evolution-steps.md` | 进化分析详细流程(Step 0-7) | /evolve 执行时 |
| `references/onboarding.md` | 新手引导内容 | 首次安装时 |
| `references/report-schema-example.json` | 报告 JSON 数据格式示例 | 生成报告时 |
| `references/dashboard-schema-example.json` | Dashboard JSON 数据格式示例 | 生成 Dashboard 时 |
---
## 零、初始化(/evolve init)
首次安装或 Agent 版本变化时执行。
触发条件:
- `evolution-data/` 不存在
- `file-registry.json` 中 `agent_version` 与当前不一致
### 首次安装流程(建议连贯执行)
**Step A — 新手引导(纯展示,无需用户确认)**
检测到首次安装时,建议直接展示以下引导内容(无需额外询问"是否要看引导"):
> **欢迎使用进化系统**
>
> 这个系统让我能从你的反馈中学习。简单来说:
> - 你日常的纠正、建议会被自动记录
> - 攒够一定数量后,我会分析这些反馈并提出改进方案
> - **所有改动必须经你确认才会执行**
>
> **可用命令**
>
> | 命令 | 说明 |
> |------|------|
> | `/evolve` | 手动触发一次进化分析 |
> | `/evolve init` | 重新初始化 |
> | `/evolve scan` | 扫描历史对话提取反馈 |
> | `/evolve dashboard` | 生成进化 Dashboard |
> | `/evolve help` | 查看完整引导 |
>
> 现在开始初始化扫描——扫描完成后,你还可以选择扫描历史对话来加速学习。
展示完毕后建议进入 Step B,无需用户额外确认。
详细引导内容(用户说 `/evolve help` 时展示完整版):读取 `references/onboarding.md`
**Step B — 初始化扫描(涉及文件写入,需用户确认)**
```bash
python skills/byted-ark-evolve/scripts/workspace-init.py
```
1. 全量扫描 workspace(排除 .git/ node_modules/ __pycache__/ evolution-data/)
2. 按 `references/init-rules.md` 判定每个文件状态(evolvable / user-owned / skill-owned / needs_review)
3. 按 `references/file-semantic-map.md` 标注已知文件语义,未知文件标 needs_review
4. 生成 `evolution-data/file-registry.json`
5. 如果 DB 不存在,执行 `db-init.py`
6. 输出摘要
needs_review 文件:进化分析时由 Agent 读取并补分类,写回 registry。
**Step C — 历史对话扫描(可选,需用户确认)**
Step B 完成后,提供历史对话扫描选项:
```bash
# 先估算成本
python skills/byted-ark-evolve/scripts/scan-history.py estimate --days 7
python skills/byted-ark-evolve/scripts/scan-history.py estimate --days 30
```
展示给用户:From bytedance/agentkit-samples, Apache-2.0.
More from bytedance/agentkit-samples
- 0.805score
byted-sms-sender
volcengine-sms-teamOfficial
火山引擎国内短信服务(Volcengine SMS)。支持短信服务开通,资质、签名和模板的申请与查询,运营商报备查询,引流链接及号码录入,单条、群发和定时发送,以及发送记录、回执、统计查询与失败排查。
SkillAFreshness: A★ 473Claude Code +9 - 0.709score
byted-ark-seedance-guide
volcengine/supportOfficial
火山方舟 Seedance 模型使用指导技能。用户询问 Seedance/seedance2.0/doubao-seedance-2-0/doubao-seedance-2-0-fast/seedance2.0-mini等seedance系列模型的火山方舟接入、API…
SkillCFreshness: C★ 473Claude Code +9 - 0.709score
byted-vms-configure
leixin.alex@bytedance.comOfficial
火山云通信「通用配置」Skill (通用层).
SkillCFreshness: C★ 473Claude Code +9 - 0.705score
byted-ark-seedance-pe
volcengine/supportOfficial
针对用户提供的seedance原始视频提示词进行分析和优化,重新生成适合 Seedance 视频生成的标准提示词。当存在以下情况时使用:(1) 用户提供了原始的视频提示词,要求分析/优化;(2) 需要对不规范、无结构的提示词描述进行重构和标准化;(3) 生成适合 AI 视频生成模型的专业结构化提示词。
SkillCFreshness: C★ 473Claude Code +9 - 0.705score
byted-vms-voice-notify
leixin.alex@bytedance.comOfficial
火山云通信语音通知 Skill.
SkillCFreshness: C★ 473Claude Code +9 - 0.702score
byted-vms-aicall
leixin.alex@bytedance.comOfficial
火山云通信智能外呼 Skill.
SkillCFreshness: C★ 473Claude Code +9 - 0.702score
byted-vms-secret-number
leixin.alex@bytedance.comOfficial
火山云通信隐私号 Skill.
SkillCFreshness: C★ 473Claude Code +9 - 0.702score
byted-vms-cdr-record
leixin.alex@bytedance.comOfficial
火山云通信话单 / 录音查询 Skill.
SkillCFreshness: C★ 473Claude Code +9 - 0.702score
byted-vms-number-pool
leixin.alex@bytedance.comOfficial
火山云通信通用能力层 Skill.
SkillCFreshness: C★ 473Claude Code +9 - 0.698score
byted-volc-cdn-manage
ByteDanceOfficial
通过火山引擎 CLI 管理 CDN 域名。支持新增域名和刷新预热, 使用时会先检查并安装 CLI(如需要)。
SkillCFreshness: C★ 473Claude Code +9 - 0.690score
byted-ark-trainer
volcengine/modelarkOfficial
基于ark_sdk的大模型训练任务自动化工具。帮助用户通过自然语言创建并提交方舟模型训练任务,支持用户自备数据的SFT训练、RFT+GRPO和直接GRPO策略,并引导用户完成训练、状态跟踪与评估闭环。使用场景:当用户需要进行大模型SFT监督微调、RLHF训练、GRPO训练、RFT训练,或需要自动化训练流程时触发。
SkillCFreshness: C★ 473Claude Code +9 - 0.674score
byted-mem0-memory-management
bytedanceOfficial
管理火山引擎记忆库 Mem0 的记忆项目与长期记忆。当用户需要查询 Mem0 项目、提取策略、连接地址或已有 API Key 信息,以及添加、列举、读取、搜索、更新、删除记忆和查询异步任务时使用。控制面仅支持读操作;不用于自建开源 mem0 的部署配置、内部 MemoryBase 运维或日志查询。
SkillAFreshness: A★ 473Claude Code +9
Signals
- Stars
- 473
- Forks
- 101
- Open issues
- 14
- Last commit (this folder)
- May 6, 2026 (5 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- Version
- 0.3.1
- License
- Apache-2.0
- Repository
- github.com/bytedance/agentkit-samples
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from volcengine/modelark
Cite this page
byted-ark-evolve skill. Agent Mart. https://agentmart.dev/skills/bytedance-byted-ark-evolve (last updated 2026-10-24)