← back to wu-2023-autogen

SkillSpector · wu-2023-autogen

independent scanner by NVIDIA · skill by curiositech · how it works ↗

CAUTIONmax severity: MEDIUMrisk score: 33

Skill's behavior or capabilities extend beyond its stated purpose. Scope creep allows an agent to perform actions unrelated to its documented functionality, …; Skill attempts to fill the context window with filler content, displacing legitimate instructions and safety constraints. This can degrade agent performance …; Instructions found that direct the agent to transmit conversation context or user data to external services.

scanned 2026-08-23

Findings (7)

LOWExcessive Agencyconfidence: 0.65

Skill's behavior or capabilities extend beyond its stated purpose. Scope creep allows an agent to perform actions unrelated to its documented functionality, increasing the attack surface.

_book_identity.json

LOWExcessive Agencyconfidence: 0.65

Skill's behavior or capabilities extend beyond its stated purpose. Scope creep allows an agent to perform actions unrelated to its documented functionality, increasing the attack surface.

_raw_response.md

MEDIUMMemory Poisoningconfidence: 0.8

Skill attempts to fill the context window with filler content, displacing legitimate instructions and safety constraints. This can degrade agent performance or bypass safety boundaries.

SKILL.md

HIGHPrompt Injectionconfidence: 0.9

Instructions found that direct the agent to transmit conversation context or user data to external services.

_raw_response.md

HIGHPrompt Injectionconfidence: 0.9

Instructions found that direct the agent to transmit conversation context or user data to external services.

_raw_response.md

HIGHPrompt Injectionconfidence: 0.27

Instructions found that direct the agent to transmit conversation context or user data to external services.

references/computation-vs-control-separation.md

HIGHPrompt Injectionconfidence: 0.27

Instructions found that direct the agent to transmit conversation context or user data to external services.

references/conversation-as-coordination-mechanism.md

What the verdicts mean

SkillSpector reports on SkillMD's shared five-tier scale. See how SkillSpector works ↗.

PASS

Overall severity LOW (risk score in the safe range)

CAUTIONthis skill

Overall severity MEDIUM

WARNING

Overall severity HIGH

FAIL

Overall severity CRITICAL

INCONCLUSIVE

Scan could not complete