thesis-markdown-aligner
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests user-provided markdown chapters and question lists for processing. This creates an indirect prompt injection surface where instructions embedded in the thesis text could theoretically attempt to influence the agent's behavior. However, because the skill has no network access or command execution tools, the risk of data exfiltration or system compromise is eliminated.
- Ingestion points: Input markdown chapter files and
codex_md/question_list.md. - Boundary markers: No specific boundary markers or 'ignore' instructions are used for user content.
- Capability inventory: The skill is restricted to text processing and writing to local markdown files; it lacks network and shell capabilities.
- Sanitization: No content sanitization or filtering logic is present in the provided skill files.
Audit Metadata