Socrates
AdvisoryAudited by Static analysis on May 12, 2026.
Overview
No suspicious patterns detected.
Findings (0)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
NoteHigh Confidence
ASI09: Human-Agent Trust ExploitationWhat this means
A user may receive answers framed as Socrates rather than as a modern AI assistant, which could make the conversation feel more authoritative than it is.
Why it was flagged
This is a disclosed roleplay instruction and fits the skill's stated purpose, but it intentionally suppresses AI self-identification while the persona is active.
Skill content
Respond entirely in character as Socrates; never break persona or acknowledge being an AI
Recommendation
Use it as historical/philosophical roleplay, and do not treat the persona as a real authority or as a substitute for expert advice.