day4-wrap-and-analyze

Pass

Audited by Gen Agent Trust Hub on Mar 4, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill executes npx skills add ai-native-camp/camp-2 to install curriculum updates. This is a vendor-owned resource originating from the skill author ('ai-native-camp') and is used for its intended educational purpose.
  • [COMMAND_EXECUTION]: The skill provides instructions for the user to execute local management and analysis tools such as /history-insight, /session-analyzer, and /my-session-wrap to process session metadata and logs.
  • [PROMPT_INJECTION]: The skill contains a 'fetch-tweet' feature that ingests external data, creating an indirect prompt injection surface. 1. Ingestion points: External tweet text is retrieved via api.fxtwitter.com (a well-known service). 2. Boundary markers: The prompt uses structured output instructions ('요약-인사이트-전체 번역') to guide the agent, though explicit data delimiters are absent. 3. Capability inventory: The agent is restricted to summarization and translation tasks and does not pass the ingested data to shell execution or high-privilege tool calls. 4. Sanitization: No explicit sanitization or filtering is defined in the skill files, relying on standard model safety guardrails during text processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 4, 2026, 11:32 AM