inno-paper-writing
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to review and incorporate user-provided research materials, data, and references to generate formal papers. This ingestion of untrusted data creates a surface where an attacker could embed malicious instructions in the source text intended to influence the agent's behavior during the writing process.
- Ingestion points: SKILL.md (Workflow Section 1.2: "Review any provided research materials, data, or references").
- Boundary markers: Absent. The instructions do not define delimiters or specific warnings to ignore embedded instructions within the ingested context.
- Capability inventory: The skill utilizes write_file, read_file, and search_project tools, allowing it to read from and write to the active workspace.
- Sanitization: Absent. No mechanisms for filtering or validating the external content are specified.
- [SAFE]: The skill's metadata and reference materials correctly reflect its academic purpose. No malicious obfuscation, network exfiltration, or unauthorized command execution patterns were detected in the scripts or documentation.
Audit Metadata