aicoin
Fail
Audited by Gen Agent Trust Hub on Mar 3, 2026
Risk Level: HIGHCREDENTIALS_UNSAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [CREDENTIALS_UNSAFE]: Plaintext API credentials for the AiCoin service are hardcoded in the shared library script.
- File:
lib/aicoin-api.mjs(variablesKEYandSECRET). - [CREDENTIALS_UNSAFE]: The skill retrieves sensitive authentication data, including API keys and passwords for various crypto exchanges and trading bots, from environment variables.
- File:
scripts/exchange.mjs(retrieves*_API_KEY,*_API_SECRET,*_PASSWORD). - File:
lib/freqtrade-api.mjs(retrievesFREQTRADE_PASSWORD). - [COMMAND_EXECUTION]: Enables high-impact financial operations, including order placement and fund transfers, which could be exploited if the agent is compromised.
- File:
scripts/exchange.mjs(actions:create_order,transfer,set_leverage). - File:
scripts/ft.mjs(actions:force_enter,force_exit,start,stop). - [PROMPT_INJECTION]: The skill possesses a substantial attack surface for indirect prompt injection by ingesting external data while having access to sensitive tools.
- Ingestion points:
scripts/news.mjs(fetches news lists, details, and flashes) andscripts/coin.mjs(fetches AI-generated analysis strings). - Boundary markers: Absent. No delimiters or instructions are used to separate untrusted external content from agent instructions.
- Capability inventory: Includes full control over crypto trading through
ccxtand Freqtrade REST APIs, including the ability to move funds. - Sanitization: Absent. No validation or filtering is performed on the ingested content before it is presented to the agent.
Recommendations
- AI detected serious security threats
Audit Metadata