Self-Evolving Harness
AdvisoryAudited by VirusTotal on Apr 19, 2026.
Overview
Type: OpenClaw Skill Name: self-evolving-harness Version: 0.0.3 The skill bundle contains only metadata and a high-level conceptual description in SKILL.md. There is no executable code, scripts, or actionable instructions that could lead to data exfiltration, unauthorized execution, or prompt injection. The content appears to be a placeholder or documentation for an agent autonomy framework.
Findings (0)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
The concept is ambitious and autonomy-focused, so users should understand what future versions or external components might do before relying on it.
The description frames the skill around end-to-end autonomous evolution and continued agent operation/evolution, which is an autonomy-oriented goal. However, the artifacts contain no code or instructions implementing persistence or autonomous background activity.
端到端自主进化 ... 即便记忆丢失，Agent 依然有序运行和进化
Install only if you are comfortable with the stated autonomy-oriented purpose, and review any future code, install steps, or permissions before enabling them.
If implemented later, autonomous knowledge storage could affect what information is retained and reused across tasks.
The description mentions autonomous knowledge accumulation, which could involve memory or retained context in a fuller implementation. The current artifact does not define any storage, retrieval, sharing, or reuse behavior.
自主知识沉淀
Before using any expanded version, check that it clearly explains what information is stored, where it is stored, and how users can inspect or delete it.