pumpfun
Pass
Audited by Gen Agent Trust Hub on Jul 9, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill relies on the execution of a system command-line tool
pumpfunto perform token launches, trades, and fee collection on the Solana network. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests untrusted data from the blockchain or user-provided metadata without explicit boundary markers or sanitization.
- Ingestion points: The skill processes external inputs such as
<mint_address>,<creator_address>, and token metadata fields (name, symbol, description) within thepumpfunCLI commands inSKILL.md. - Boundary markers: The instructions do not define delimiters or provide warnings to the agent to ignore potentially malicious instructions embedded in token metadata.
- Capability inventory: The skill utilizes subprocess execution via the
pumpfuntool to perform financial transactions and read network state. - Sanitization: There is no evidence of input validation or escaping for the external data processed by the tool.
Audit Metadata