unlock-monitor
Pass
Audited by Gen Agent Trust Hub on Aug 25, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: Indirect Prompt Injection Surface. The skill fetches and processes data from external, third-party cryptocurrency websites and search results. This data is used to generate analysis and notifications, creating a vector where an attacker could influence the agent's behavior by manipulating content (e.g., token names or descriptions) on the source platforms.
- Ingestion points: Step 1 uses WebSearch and WebFetch to collect data from multiple crypto-related domains including Tokenomist, DeFi Llama, and CoinGecko.
- Boundary markers: Absent. The instructions do not define delimiters or specific "ignore embedded instructions" headers when processing the external data.
- Capability inventory: The skill utilizes
contents:writepermissions for persistence in thememory/directory and performs network operations via WebFetch. - Sanitization: Absent. There is no specified logic for sanitizing or validating external strings before they are used in market-read classification or notification outputs.
Audit Metadata