writing-style
Warn
Audited by Gen Agent Trust Hub on May 14, 2026
Risk Level: MEDIUMDATA_EXFILTRATIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [DATA_EXFILTRATION]: The skill instructs the agent to read content from local file paths provided by the user (SKILL.md, Section 1). This capability could be abused to access sensitive system files or credentials if the agent is misled into 'analyzing' them for writing style.
- [EXTERNAL_DOWNLOADS]: The skill is designed to fetch and analyze content from remote GitHub repositories based on user-provided URLs (SKILL.md, Section 1).
- [PROMPT_INJECTION]: There is a risk of indirect prompt injection as the skill ingests untrusted data from various sources (pasted text, local files, and remote repositories) and passes the resulting analysis to a subsequent tool (
/spencergo:writing-outline). - Ingestion points: User-pasted text, local file paths, and GitHub repository URLs (SKILL.md, Section 1).
- Boundary markers: Not present.
- Capability inventory: File system reading, network fetching of repository data, and tool invocation using the processed context.
- Sanitization: Not present.
Audit Metadata