tw-research-synchronizer
Pass
Audited by Gen Agent Trust Hub on May 1, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill functions as a questionnaire that stores user responses in a configuration file used as context for other skills. This creates an attack surface for indirect prompt injection if a user provides malicious instructions instead of research data.
- Ingestion points: User responses to survey questions Q1-Q18 in SKILL.md.
- Boundary markers: The generated 'researcher-profile.md' does not employ delimiters or 'ignore instructions' markers to isolate user-provided data.
- Capability inventory: The skill uses 'Bash', 'Read', and 'Write' for local file system operations and a 'Notion' tool for data backup.
- Sanitization: No input validation or character escaping is applied to user responses before they are written to the configuration file.
Audit Metadata