add-gchat
Pass
Audited by Gen Agent Trust Hub on Apr 26, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches code and dependencies from external sources as part of the installation process.
- Fetches source files from a remote Git branch (
origin/channels) usinggit fetchandgit show. - Installs the
@chat-adapter/gchat@4.26.0package from the npm registry. - [COMMAND_EXECUTION]: The skill requires the execution of several shell commands to configure the environment and build the application.
- Runs
pnpm installandpnpm run buildduring setup. - Uses file operations (
mkdir,cp) to sync environment configurations to a data directory. - [INDIRECT_PROMPT_INJECTION]: The skill enables a communication channel that ingests data from Google Chat spaces, creating a surface for indirect prompt injection where external messages could influence agent behavior.
- Ingestion points: Incoming messages and mentions from Google Chat spaces.
- Boundary markers: None explicitly mentioned; the skill relies on the underlying framework's handling of message data.
- Capability inventory: Enables the agent to respond to messages and potentially interact with framework tools via the Google Chat interface.
- Sanitization: The instructions do not specify any unique sanitization steps for the incoming message stream beyond standard platform behaviors.
Audit Metadata