reverse-character-cards
Pass
Audited by Gen Agent Trust Hub on Aug 8, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill performs character reverse-engineering based on provided manuscript text. It contains explicit instructions to never modify the author's prose, acting solely as a documentation tool.
- [SAFE]: The use of the Model Context Protocol (MCP) to fetch chapters and update the 'Map' (a character database) is a standard feature of the calliope-editor ecosystem and does not involve unauthorized data access or exfiltration.
- [SAFE]: The skill implements a structured evidence chain, tagging information by provenance (narrated, claimed, inferred), which serves as a functional boundary when processing untrusted manuscript data.
- [SAFE]: No obfuscation, hardcoded credentials, remote code execution, or persistence mechanisms were detected in the skill's instructions or metadata.
Audit Metadata