psychic-distance
Pass
Audited by Gen Agent Trust Hub on Aug 8, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is designed to ingest and evaluate user-submitted prose, which introduces a surface for indirect prompt injection if a user includes malicious instructions within their writing samples.
- Ingestion points: User-supplied text passages pasted into the conversation and manuscript data accessed via the MCP connector.
- Boundary markers: The instructions do not define specific delimiters to separate user text from agent instructions, nor do they instruct the agent to ignore embedded commands in the prose.
- Capability inventory: The skill is restricted to reading text and generating feedback. It uses a read-only MCP connector and lacks capabilities for file system modification, shell command execution, or unauthorized network operations.
- Sanitization: There is no evidence of sanitization or input validation performed on the ingested prose.
- [SAFE]: The skill references a documentation URL (writecalliope.ink/docs/mcp) for the author's Model Context Protocol (MCP) connector. This is a standard resource for the skill's primary purpose and does not represent a security risk.
Audit Metadata