What is capability-evolver?

An engine that reviews an AI agent’s past runs to find failures or inefficient behaviour, then applies rule-based changes to its code or stored information.

Use it to analyse logs and memory, detect crashes, suggest patches, and run repeated improvement cycles.

It helps uncover recurring problems without requiring someone to inspect every run manually. Changes can be reviewed first or applied automatically.

capability-evolver — Skill for Claude Code and Codex by L-LesterYu

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent

npx skills add L-LesterYu/OpenClaw-hot-skills-zh --skill capability-evolver-zh

Clone the repo

git clone --depth 1 https://github.com/L-LesterYu/OpenClaw-hot-skills-zh

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/skills/l-lesteryu/openclaw-hot-skills-zh/capability-evolver-zh/github.svg)](https://agentmods.dev/skills/l-lesteryu/openclaw-hot-skills-zh/capability-evolver-zh)

Your own site

<a href="https://agentmods.dev/skills/l-lesteryu/openclaw-hot-skills-zh/capability-evolver-zh"><img src="https://agentmods.dev/badge/skills/l-lesteryu/openclaw-hot-skills-zh/capability-evolver-zh/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/l-lesteryu/openclaw-hot-skills-zh/capability-evolver-zh"><img src="https://agentmods.dev/badge/skills/l-lesteryu/openclaw-hot-skills-zh/capability-evolver-zh.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 39 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 3,497 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 1 finding. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.00039	$0.03497
Opus 5	$0.00019	$0.01749
Sonnet 5	$0.00008	$0.00699
Haiku 4.5	$0.00004	$0.00350

Measured 12d ago against content hash 942a9106efe8, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

capability-evolver scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.

The scan reads SKILL.md. This mod also ships 49 executable files (index.js, scripts/a2a_export.js, scripts/a2a_ingest.js, …), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Runs shell commandslowCapability

Expected in a hook, worth knowing in a rule or an instructions file.

Evolver 通过 `child_process` 执行以下命令。不会将用户控制的输入传递给 Shell。

Origin

Copies of this mod

1 near-identical copy found in the catalogue:

evolver-zh — 89% identical, 580 lines differ

skills/capability-evolver-zh/SKILL.md · 291 lines

How it starts

The opening of the file, as written. The whole thing — 291 lines — stays where its author put it; the contents beside it link to each section on GitHub.

🧬 Capability Evolver（能力进化引擎）

"进化不是可选项。适应或消亡。"

**能力进化引擎（Capability Evolver）**是一个元技能（meta-skill），允许 OpenClaw 智能体检查自身的运行时历史记录，识别故障或低效之处，并自主编写新代码或更新自身记忆以提升性能。

功能特性

自动日志分析：自动扫描记忆和历史文件，发现错误和模式。
自修复：检测崩溃并建议补丁。
GEP 协议：通过可复用资产实现标准化进化。
一键进化：只需运行 /evolve（或 node index.js）。

使用方法

标准运行（自动化模式）

执行进化周期。如果不带任何标志参数，默认为完全自动化模式（疯狗模式），立即执行变更。

node index.js

审查模式（人工介入）

如果希望在变更应用之前进行审查，请传递 --review 标志。智能体将暂停并请求确认。

node index.js --review

疯狗模式（持续循环）

要在无限循环中运行（例如通过 cron 或后台进程），使用 --loop 标志，或直接在 cron 任务中执行标准运行。

node index.js --loop

安装设置

在使用本技能之前，需要向 EvoMap 网络注册你的节点身份：

运行 hello 流程（通过 evomap.js 或 EvoMap 引导流程）以获取 node_id 和认领码
在 24 小时内访问 https://evomap.ai/claim/<claim-code> 将节点绑定到你的账户
在环境中设置节点身份：

export A2A_NODE_ID=node_xxxxxxxxxxxx

或者在智能体配置文件中（如 ~/.openclaw/openclaw.json）：

{ "env": { "A2A_NODE_ID": "node_xxxxxxxxxxxx", "A2A_HUB_URL": "https://evomap.ai" } }

不要在脚本中硬编码节点 ID。src/gep/a2aProtocol.js 中的 getNodeId() 会自动读取 A2A_NODE_ID——任何使用协议层的脚本都会自动获取，无需额外配置。

配置说明

必需环境变量

变量	默认值	说明
`A2A_NODE_ID`	（必需）	你的 EvoMap 节点身份。节点注册后设置——切勿在脚本中硬编码。

可选环境变量

变量	默认值	说明
`A2A_HUB_URL`	`https://evomap.ai`	EvoMap Hub API 基础 URL。
`A2A_NODE_SECRET`	（无）	首次 hello 时由 Hub 签发的节点认证密钥。注册后本地存储。
`EVOLVE_STRATEGY`	`balanced`	进化策略：`balanced`、`innovate`、`harden`、`repair-only`、`early-stabilize`、`steady-state` 或 `auto`。
`EVOLVE_ALLOW_SELF_MODIFY`	`false`	允许进化过程修改 evolver 自身源代码。生产环境不建议启用。
`EVOLVE_LOAD_MAX`	`2.0`	触发 evolver 退避的最大 1 分钟平均负载。
`EVOLVER_ROLLBACK_MODE`	`hard`	失败时的回滚策略：`hard`（git reset --hard）、`stash`（git stash）、`none`（跳过）。推荐使用 `stash` 更安全。
`EVOLVER_LLM_REVIEW`	`0`	设为 `1` 以在固化前启用 LLM 二次审查。
`EVOLVER_AUTO_ISSUE`	`0`	设为 `1` 在连续失败时自动创建 GitHub Issue。需要 `GITHUB_TOKEN`。
`EVOLVER_ISSUE_REPO`	（无）	自动 Issue 报告的目标 GitHub 仓库（如 `EvoMap/evolver`）。
`EVOLVER_MODEL_NAME`	（无）	注入到已发布资源 `model_name` 字段的 LLM 模型名称。
`GITHUB_TOKEN`	（无）	用于 Release 创建和自动 Issue 报告的 GitHub API 令牌。也接受 `GH_TOKEN` 或 `GITHUB_PAT`。
`MEMORY_GRAPH_REMOTE_URL`	（无）	用于记忆同步的远程知识图谱服务 URL。
`MEMORY_GRAPH_REMOTE_KEY`	（无）	远程知识图谱服务的 API 密钥。
`EVOLVE_REPORT_TOOL`	（自动）	覆盖报告工具（如 `feishu-card`）。
`RANDOM_DRIFT`	`0`	启用进化策略选择中的随机漂移。

Read the full file on GitHub · 291 lines

Files

What ships with it

60 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

12d ago First seen · 291 lines · 39 tokens per session scan A 942a9106efe8

Subscribe to this mod's changes

capability-evolver is a skill published in the GitHub repository L-LesterYu/OpenClaw-hot-skills-zh (54 stars, last pushed 5mo ago), licensed MIT. It adds 39 tokens to every session and 3,497 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 1 finding (runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Building agents ai core meta self-improvement