gp-cli-skill
Warn
Audited by Gen Agent Trust Hub on Apr 6, 2026
Risk Level: MEDIUMCREDENTIALS_UNSAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [CREDENTIALS_UNSAFE]: The skill instructions direct the agent to access highly sensitive Solana private key files (e.g.,
~/.config/solana/id.json) to register wallets with the CLI tool. - [EXTERNAL_DOWNLOADS]: The skill requires the installation of the
@graveyardprotocol/gp-clipackage from the NPM registry to provide its core functionality. - [COMMAND_EXECUTION]: The skill relies on the execution of shell commands via the
gpbinary for all operations, including adding wallets, closing accounts, and claiming rewards. - [DATA_EXFILTRATION]: The CLI communicates with a remote backend at
https://api.graveyardprotocol.ioto process transactions. When combined with the skill's access to local private key files, this creates a potential vector for data exposure, despite the skill's claims of local signing. - [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection from external data sources. 1. Ingestion points: The agent processes scan summaries and epoch statistics fetched from the Graveyard Protocol API. 2. Boundary markers: There are no explicit markers or safety instructions used to delimit untrusted data from the backend. 3. Capability inventory: The skill has the ability to execute arbitrary shell commands via the
gptool and write files to the local disk. 4. Sanitization: The instructions do not specify any validation or sanitization of the data received from the remote API before it is processed by the agent.
Audit Metadata