Scale Frameworks
PassAudited by VirusTotal on May 11, 2026.
Overview
Type: OpenClaw Skill Name: scale Version: 1.0.0 The skill is designed as an advisory and planning tool, explicitly stating it does not perform infrastructure changes or network requests without user confirmation. The `SKILL.md` and `setup.md` files contain clear instructions for the AI agent to store data locally within `~/scale/`, to redact sensitive identifiers, and explicitly states 'None by default from this skill itself' for data leaving the machine. It also disclaims storing secrets or modifying files outside its designated memory path. The prompt injection elements are used to guide the agent's persona and enforce safe data handling practices, not to subvert security or exfiltrate data.
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.
There is limited code-execution risk from the artifacts, but users have less provenance information about who authored the guidance.
The publisher source/provenance is not specified, but the provided package contains only text instructions and no executable install mechanism.
Source: unknown; No install spec — this is an instruction-only skill.
Review the instruction files before use and prefer skills from publishers you trust, especially for business or infrastructure planning.
Outdated, inaccurate, or sensitive business context could persist locally and shape later recommendations.
The skill intentionally stores durable local planning context that can influence future scaling advice.
Persist only reusable scaling intelligence in `~/scale/memory.md`: - Integration preference - Bottleneck patterns by domain - Interventions that worked and why - Guardrails that prevented regressions
Keep the memory file concise, avoid secrets or sensitive identifiers, and periodically review or delete files under ~/scale/ if the context changes.