cdp
Warn
Audited by Gen Agent Trust Hub on Sep 2, 2026
Risk Level: MEDIUMDYNAMIC_EXECUTIONCOMMAND_EXECUTIONPERSISTENCEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [DYNAMIC_EXECUTION]: The skill's core functionality relies on a Node.js HTTP server (
sdk/repl.ts) that accepts raw JavaScript code via aPOST /evalendpoint and executes it usingeval(). - While the server binds to
127.0.0.1, any local process can interact with it on the default port 9876 to execute arbitrary code. - The skill also uses dynamic
import()insdk/helpers.tsto load site-specific tools from thelearnings/directory based on manifest files. - [COMMAND_EXECUTION]: The harness provides an environment where the agent can execute any Node.js logic, including interacting with the host filesystem and spawning child processes.
- Interaction recipes in
interaction-skills/document usingnode:fsfor writing files andosascriptviaexecFilefor UI automation tasks. - [PERSISTENCE]: The
scripts/setupscript automatically modifies user shell configuration files (.zshrc,.bashrc, or.profile) to ensure the CLI tool is available in future sessions. - It appends a command to export a new directory into the user's
PATHenvironment variable. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process content from any website the agent navigates to, creating a significant surface for indirect injection.
- Ingestion points: Untrusted data enters the context via accessibility tree snapshots (
axViewinsdk/axview.ts) and async page signals (drainSignalsinsdk/helpers.ts). - Boundary markers: The
axViewhelper produces structured text but lacks explicit markers to prevent the agent from obeying instructions embedded in the web content. - Capability inventory: The skill possesses high-privilege capabilities including cookie manipulation, credential scrubbing, and arbitrary browser interaction.
- Sanitization: Minimal sanitization is performed on page content before it is passed to the LLM.
- [EXTERNAL_DOWNLOADS]: The skill uses
curlin its bash CLI and documentedfetchpatterns in the interaction guides to interact with local and remote resources. - [DATA_EXPOSURE]: The tool contains logic in
sdk/session.tsto scan sensitive OS directories for browser profile data to detect running Chromium instances and read their debugging ports.
Audit Metadata