Book Writing
PassAudited by VirusTotal on May 11, 2026.
Overview
Type: OpenClaw Skill Name: book-writing Version: 1.0.0 The OpenClaw AgentSkills bundle for 'book-writing' is classified as benign. All files, including the SKILL.md instructions for the AI agent, consistently describe a book writing assistance tool. File system access is explicitly limited to the `~/book-writing/` directory for local memory management, and the skill explicitly states it does not send data to external APIs or access files outside its designated directory. There are no instructions for malicious actions, data exfiltration, unauthorized execution, or persistence mechanisms, and no external binaries are required. The prompt injection surface (SKILL.md, setup.md) only contains instructions relevant to the stated purpose of assisting with book writing.
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.
Project details, voice preferences, chapter plans, and revision notes may be reused in later sessions, so inaccurate or overly personal entries could influence future outputs.
The skill intentionally creates persistent local memory for book projects. This is disclosed and purpose-aligned, but persistent notes can affect future writing sessions and may contain sensitive manuscript context.
When local memory is enabled: - Create `~/book-writing/memory.md` from `memory-template.md`. - Update `last` after each meaningful session. - Keep summaries short and decision-focused. - Never store secrets, credentials, or unrelated personal data.
Enable local memory only if useful, periodically review or edit `~/book-writing/memory.md`, and avoid storing secrets or unrelated personal information there.