← 返回 Skills 市场

Evolution Toolkit

Name: Evolution Toolkit
Author: ergopitrez

作者 ergo · GitHub ↗ · v1.0.0 · MIT-0

cross-platform ⚠ suspicious

总下载

当前安装

版本数

在 OpenClaw 中安装

/install evolution-toolkit-build

功能描述

Provides tools for agent self-improvement including session handoffs, reasoning analysis, contradiction scanning, prediction logging, playbook optimization,...

使用说明 (SKILL.md)

Evolution Toolkit

Skill by Ergo | 2026-03-24 | Portable toolkit for agent self-improvement across workspaces Status: ✅ Verified

How to use: Set EVOLUTION_TOOLKIT_WORKSPACE to the target workspace, then run the script that matches the cognitive task. Load protocols/session-continuity.md or protocols/thinking-partner.md when the task is about handoffs or problem framing.

Triggers

Use this skill when the request is about any of these:

Session end, handoff, continuity, preserving context between runs
"How am I reasoning?" or "compare these sessions/documents"
Contradictions, drift, conflicting instructions, stale guidance
Prediction logging, confidence calibration, decision audits
Improving a prompt/playbook through repeated eval loops
Switching into Socratic questioning instead of direct advice
Cross-session consistency, identity drift, recurring themes

Workspace

Export a writable workspace before running any script that writes state:

export EVOLUTION_TOOLKIT_WORKSPACE=/path/to/workspace

Expected layout:

memory/
memory/imprints/ for session imprints
memory/research/ for coherence reports
CURRENT.md if you want session-imprint context
memory/prediction-log.md if you want prediction logging

Scripts

scripts/session-imprint.js

Interactive session-end handoff.
Use --read, --list, or --diff to inspect existing imprints.

scripts/cognitive-fingerprint.js

Measures reasoning style across 14 dimensions.
Useful for one file, today's log, all imprints, or historical comparisons.

scripts/contradiction-scanner.js

Scans guidance files in the workspace for conflicting directives, stale references, and drift.

scripts/predict.js

Logs predictions and audits calibration.
Requires memory/prediction-log.md with ## Log and ## Calibration sections.

scripts/skill-optimizer.js

Runs an iterative generate -> evaluate -> improve loop for a configurable playbook.
Requires a config file; see config.example.json.

scripts/socratic-mode.js

Classifies a problem into thinking phase and outputs friction-injecting questions.

scripts/session-coherence.js

Analyzes daily logs for persistent themes, energy, and drift.
Writes a report to memory/research/ by default.

Quick Commands

node scripts/session-imprint.js
node scripts/cognitive-fingerprint.js --daily
node scripts/contradiction-scanner.js --verbose
node scripts/predict.js add
node scripts/socratic-mode.js "Should I launch now or keep polishing?"
node scripts/session-coherence.js --days 14 --portrait
node scripts/skill-optimizer.js --config ./config.json --skill customer-support --iterations 3

Protocols

Read these only when relevant:

protocols/session-continuity.md: how to end and resume sessions cleanly
protocols/thinking-partner.md: how to add useful friction instead of reflexive answers

Notes

Write-capable scripts exit early with a clear warning if the target workspace is not writable.
skill-optimizer.js is intentionally config-driven so the package stays product-neutral.
The toolkit does not ship credentials. API keys must come from env vars or your own workspace secrets.

安全使用建议

This toolkit appears to do what it says: it reads workspace files, analyzes reasoning and consistency, logs predictions, and can optimize playbooks — and the optimizer may use an LLM API if you supply a key. Before installing or running: 1) Inspect skill-optimizer.js and socratic-mode.js for any network calls (search for fetch/http/https/request/axios/openai/gemini/googleapis or direct URLs). 2) Run the scripts first against a disposable, read-only or isolated workspace to observe behavior; many scripts check write access and will exit if workspace is not writable. 3) If you provide an API key, assume text from your workspace may be sent to that provider — only supply keys with least privilege and for accounts you control. 4) Because the package source/homepage is not provided, prefer running it locally (not enabling always:true) and review all included files for unexpected endpoints/telemetry. If you want higher assurance, provide the missing full contents of skill-optimizer.js and socratic-mode.js for review; that would likely change the confidence to high and could move the verdict to benign if no unexpected external transmission is found.

功能分析

Type: OpenClaw Skill Name: evolution-toolkit-build Version: 1.0.0 The Evolution Toolkit is a collection of Node.js utilities designed for agent session continuity and self-improvement. While the toolkit includes high-risk capabilities such as reading workspace secrets and communicating with the Google Gemini API (in `scripts/skill-optimizer.js`), these behaviors are explicitly documented and aligned with the stated purpose of optimizing playbooks and analyzing reasoning. No evidence of malicious intent, unauthorized data exfiltration, or harmful prompt injection was found across the scripts or markdown protocols.

能力评估

✓ Purpose & Capability

The name/description (agent self-improvement: session handoffs, contradiction scanning, prediction logging, playbook optimization, Socratic questioning, coherence analysis) matches the included scripts and protocols. The repo contains CLI Node scripts that read and write a workspace and implement the advertised features. The only external capability hinted at is optional LLM usage by the skill-optimizer (README and config.example reference GEMINI_API_KEY/GOOGLE_API_KEY), which is coherent with a generate->evaluate->improve optimizer.

ℹ Instruction Scope

SKILL.md instructs the agent/operator to set EVOLUTION_TOOLKIT_WORKSPACE and run scripts that read many workspace files (memory/, guidance files, CURRENT.md, etc.) and write imprints/reports. This is consistent with its function, but the scripts scan and can aggregate large amounts of workspace content — review what they send externally. Two significant script files (skill-optimizer.js and socratic-mode.js) were present but their full contents were truncated in the material provided; if they call LLM APIs or remote endpoints they could transmit workspace data. Scripts include clear write-access checks and exit early if workspace not writable.

✓ Install Mechanism

There is no install spec (instruction-only install). No external archives or installers are fetched by the skill metadata. The code is shipped as local scripts to be run with node; there is no automated download-from-URL step in the registry metadata. This is low-install risk, but running the included scripts will place/read files in the chosen workspace directory.

ℹ Credentials

The declared skill requires no environment variables or credentials. The README and config.example explicitly mention optional API keys (GEMINI_API_KEY, GOOGLE_API_KEY or workspace .secrets) for the skill-optimizer — that is proportionate to an optimizer that can call external LLM services. No unrelated secrets or system credentials are requested. However, if you provide an API key, expect the optimizer to use it and potentially send workspace text to the provider; confirm what data is transmitted before supplying keys.

✓ Persistence & Privilege

always is false and the skill does not request any elevated agent/system privileges. Scripts read/write only within the EVOLUTION_TOOLKIT_WORKSPACE and perform no modifications to other skills or system-wide configs. Autonomous invocation (model invocation enabled) is default but not combined here with always:true or extra credential access.

如何使用

确保已安装 OpenClaw（本地或 Docker 部署）
在对话框中输入安装命令：/install evolution-toolkit-build
安装完成后，直接呼叫该 Skill 的名称或使用 /evolution-toolkit-build 触发
根据 Skill 的参数说明提供必要输入，即可获得结构化输出

版本历史

v1.0.0

v1.0 — 7 tools for AI agent self-improvement: session imprints, cognitive fingerprinting, contradiction scanning, prediction calibration, Socratic thinking partner, session coherence.

元数据

Slug evolution-toolkit-build

版本 1.0.0

许可证 MIT-0

累计安装 0

当前安装数 0

历史版本数 1

常见问题