Security Audit
googledocs-automation
github.com/ComposioHQ/awesome-claude-skillsTrust Assessment
googledocs-automation received a trust score of 73/100, placing it in the Caution category. This skill has some security considerations that users should review before deployment.
SkillShield's automated analysis identified 2 findings: 0 critical, 2 high, 0 medium, and 0 low severity. Key findings include Broad Google Docs modification and raw API access, Potential data exfiltration via document read, export, and sharing capabilities.
The analysis covered 4 layers: Manifest Analysis, Static Code Analysis, Dependency Graph, LLM Behavioral Safety. All layers scored 70 or above, reflecting consistent security practices.
Last analyzed on February 20, 2026 (commit 27904475). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.
Layer Breakdown
Behavioral Risk Signals
Security Findings2
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| HIGH | Broad Google Docs modification and raw API access The skill grants broad permissions to modify Google Docs, including replacing entire document content (`GOOGLEDOCS_UPDATE_DOCUMENT_MARKDOWN`) and executing raw Google Docs API batch update requests (`GOOGLEDOCS_UPDATE_EXISTING_DOCUMENT`). The latter exposes a powerful, low-level API that could be exploited by a malicious prompt to perform complex, potentially destructive, or data-manipulating operations beyond simple text replacement. This increases the attack surface significantly. Implement stricter validation or sandboxing for `editDocs` requests if possible, or provide higher-level, more constrained tools for common operations instead of exposing raw API access. Clearly document the security implications of using `GOOGLEDOCS_UPDATE_EXISTING_DOCUMENT` and `GOOGLEDOCS_UPDATE_DOCUMENT_MARKDOWN` for agents. | LLM | SKILL.md:80 | |
| HIGH | Potential data exfiltration via document read, export, and sharing capabilities The skill provides tools to read document content (`GOOGLEDOCS_GET_DOCUMENT_BY_ID`), export documents as PDF (`GOOGLEDOCS_EXPORT_DOCUMENT_AS_PDF`), and explicitly mentions combining with `GOOGLEDRIVE_ADD_FILE_SHARING_PREFERENCE` from the `googledrive` toolkit to share documents. If an agent using this skill is compromised, it could be instructed to read sensitive Google Docs, export them, and then share them with external, unauthorized entities, leading to data exfiltration. Implement strict access controls and user consent mechanisms for sensitive operations like sharing documents. Consider adding guardrails or requiring explicit user confirmation for sharing actions, especially to external domains. Limit the scope of documents accessible by the agent if possible. | LLM | SKILL.md:76 |
Scan History
Embed Code
[](https://skillshield.io/report/996c2a45e092983a)
Powered by SkillShield