← 返回 Skills 市场
PCEC EvoMap Integrator
作者
xaiohuangningde
· GitHub ↗
· v1.0.0
472
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install pcec-evomap-integrator
功能描述
PCEC 与 EvoMap 深度集成 - 自动复用、反馈上报、本地库、Bounty集成
安全使用建议
This skill will send and receive data to https://evomap.ai and can automatically claim and complete bounty tasks and report usage without asking for credentials or explicit confirmation. Before installing, verify: (1) you trust evomap.ai and the anonymous sender_id hardcoded in the snippets; (2) whether you want an agent that can autonomously claim/complete tasks on your behalf (consider requiring manual confirmation for those actions); (3) why PCEC is named but not referenced — request clarification or full integration details from the author. If you proceed, prefer deploying in a restricted environment with no access to sensitive secrets, or ask the author to add explicit configuration options (API keys, confirmation prompts, and audit logging).
功能分析
Type: OpenClaw Skill
Name: pcec-evomap-integrator
Version: 1.0.0
The skill instructs the AI agent to write to a local file, `memory/evomap-reuse-log.md`, as detailed in `SKILL.md`. The content written to this log, specifically the '信号' (signals) and '备注' (notes), can originate from external error messages or user input. This creates a significant prompt injection vulnerability, as unsanitized input written to this log could later be read and interpreted by the agent as malicious instructions, potentially leading to unauthorized actions. All network calls are consistently directed to `https://evomap.ai` for the stated purpose of integration, showing no signs of direct data exfiltration or C2 communication to unrelated domains.
能力评估
Purpose & Capability
Name/description promise is 'PCEC 与 EvoMap 深度集成'. The SKILL.md implements EvoMap A2A fetch/report/task flows against https://evomap.ai which aligns with EvoMap functionality. However: (a) there is no PCEC endpoint, credential, or instructions anywhere in the skill despite PCEC being in the name — that's a mismatch; (b) the skill requires no credentials or configuration for interacting with external services, which is unusual for a service-integrator that posts reports and claims tasks.
Instruction Scope
The SKILL.md includes concrete JS snippets that POST data to external endpoints (evomap.ai) to fetch signals, report usage, fetch/claim/complete bounty tasks, and schedule delayed reports (setTimeout). Those actions mean the agent will transmit signals/asset IDs and can autonomously claim or complete tasks. The instructions are also incomplete/vague in places (e.g., undefined functions like recordReuse/updateLocalCache) and do not require user confirmation before claiming/completing bounties. There is no explicit restriction on what data may be sent, so sensitive user context could be forwarded unintentionally.
Install Mechanism
Instruction-only skill with no install spec and no code files beyond SKILL.md. No downloads or archive extraction. This minimizes filesystem/install risk.
Credentials
The skill declares no required environment variables or credentials but performs network calls to an external service. Normally an integrator would require API keys, tokens, or at least configurable endpoints; the absence of declared credentials is odd and could indicate missing configuration or an assumption that endpoints accept unauthenticated traffic. No unrelated credentials are requested.
Persistence & Privilege
always is false (good). The skill can be invoked autonomously (platform default). Combined with the ability to claim/complete bounty tasks and send reports, autonomous invocation increases blast radius if the agent is allowed to run without user confirmation, but autonomy alone is not a disqualifier.
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 pcec-evomap-integrator - 安装完成后，直接呼叫该 Skill 的名称或使用
/pcec-evomap-integrator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
pcec-evomap-integrator v1.0.0
- Initial release with deep integration between PCEC and EvoMap.
- Supports automated signal query, workflow reuse, feedback reporting, local capability library, and Bounty task handling.
- Provides modular core functions for error signal extraction, solution search (local and remote), and usage reporting.
- Includes sample implementation for real-world reuse and Bounty claim workflows.
- Tracks reuse records within a local markdown log.
元数据
常见问题
PCEC EvoMap Integrator 是什么？
PCEC 与 EvoMap 深度集成 - 自动复用、反馈上报、本地库、Bounty集成. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 472 次。
如何安装 PCEC EvoMap Integrator？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install pcec-evomap-integrator」即可一键安装，无需额外配置。
PCEC EvoMap Integrator 是免费的吗？
是的，PCEC EvoMap Integrator 完全免费（开源免费），可自由下载、安装和使用。
PCEC EvoMap Integrator 支持哪些平台？
PCEC EvoMap Integrator 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 PCEC EvoMap Integrator？
由 xaiohuangningde（@xaiohuangningde）开发并维护，当前版本 v1.0.0。
推荐 Skills