developmental-cell
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: A comprehensive review of the skill instructions, metadata, and decision logic reveals no indicators of prompt injection, credential exfiltration, or unauthorized command execution. The skill functions solely as a cognitive guide for journal selection.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted third-party data in the form of scientific manuscripts.
- Ingestion points: The agent is prompted to evaluate manuscripts/studies against the journal's scope and evidence bar (SKILL.md).
- Boundary markers: The skill does not define specific delimiters to separate user manuscript data from the agent's instructions.
- Capability inventory: The skill does not request or utilize any dangerous capabilities such as file system writing, network requests (curl/wget), or shell command execution.
- Sanitization: No explicit sanitization or filtering of input data is defined, but the lack of exploitable tools significantly limits the risk profile.
Audit Metadata