connect-project

Pass

Audited by Gen Agent Trust Hub on Nov 3, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes git remote get-url origin to identify the repository's owner and name from the git configuration. This is a standard and safe operation used to establish project context.
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it processes external data from git remotes and user-provided URLs.
  • Ingestion points: Git remote command output and URLs provided by the user for project identification.
  • Boundary markers: The skill does not define explicit delimiters for the ingested data, but it expects specific structured formats.
  • Capability inventory: The skill is capable of calling the chromatic_query tool and performing file-write operations on chromatic.config.json, AGENTS.md, and CLAUDE.md.
  • Sanitization: Security risks are mitigated by strict validation of the project ID (requiring a 24-character hex string), explicit warnings against using sensitive project tokens, and requiring user consent before updating documentation files.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 3, 2026, 04:28 PM