Security audit
Security checks for vulnerabilities and agentic risk
This skill is a markdown-only narrative analysis framework and does not request sensitive access, execute code, or persist changes.
This appears safe to install as a writing and media-analysis aid. Be aware that it may steer broad canon or contradiction questions toward its specific 'unreliable teleplay' interpretation, so users who want strict continuity analysis may need to ask for that explicitly.
Skill contains instructions that could directly expose system prompts, internal rules, or hidden instructions to users or external parties.
- **Not retcon** — no version is rewritten or declared "correct"
- **Not canon hierarchy** — no "primary" vs "secondary" canon; each work has local authority
## Output Guidelines
When applying UTL in analysis:
- Name the event anchors explicitly
Line L03 says the skill triggers on specific phrases but also on "any request to reconcile contradictory accounts of shared characters, events, or worlds." That catch-all condition is broad and lacks negative examples or contextual constraints, which could cause unintended invocation for ordinary discussion of inconsistencies outside the intended narrative-analysis niche.
No suspicious patterns detected.