pp-gladly
Pass
Audited by Gen Agent Trust Hub on Jul 20, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill facilitates the installation of the 'gladly-pp-cli' tool and an MCP server via 'npx' and 'go install' from sources associated with the author (github.com/mvanhorn and @mvanhorn on npm). These are documented as vendor-provided resources.
- [DATA_EXFILTRATION]: The CLI tool supports a '--deliver webhook:' flag, allowing users or agents to POST command output to any arbitrary URL. It also features a 'feedback' command that can send data to a remote endpoint if the 'GLADLY_FEEDBACK_ENDPOINT' environment variable is set.
- [PROMPT_INJECTION]: The skill uses the '$ARGUMENTS' variable to construct shell commands and processes data from the Gladly API, creating a surface for indirect prompt injection. * Ingestion points: The '$ARGUMENTS' input and responses from the Gladly API. * Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the interpolation logic. * Capability inventory: The skill has access to the 'Bash' tool, allowing for subprocess execution, network operations, and file system modifications. * Sanitization: There is no evidence of input validation or escaping before data is passed to the shell or processed by the agent.
- [COMMAND_EXECUTION]: The skill relies on the 'Bash' tool to execute the 'gladly-pp-cli' binary and manage installation and configuration tasks.
Audit Metadata