The Agent Skills Directory

[DATA_EXFILTRATION]: The skill reads sensitive credentials from ~/.qodo/config.json, including GERRIT_HTTP_PASSWORD, BB_APP_PASSWORD, and AZURE_DEVOPS_EXT_PAT. This is used for authenticating with git provider APIs.
[COMMAND_EXECUTION]: The skill makes extensive use of system commands including git (for branch management and commits), curl (for API interactions), and provider CLIs like gh, glab, and az.
[EXTERNAL_DOWNLOADS]: The skill provides instructions to download a commit-msg hook from a user-configured Gerrit server URL and set it as executable.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through its core functionality of following "agent prompts" from PR comments.
Ingestion points: Fetches untrusted comment data via provider APIs (e.g., gh api, glab mr view, Bitbucket/Gerrit REST endpoints).
Boundary markers: No delimiters or isolation instructions are provided to distinguish between the intended Qodo agent prompt and potentially malicious injected content in comments.
Capability inventory: The agent has full access to modify files using the Edit tool, commit changes to the repository, and perform network operations.
Sanitization: The skill lacks validation or sanitization mechanisms to ensure the instructions in the comment body are legitimate and safe before they are followed.

qodo-pr-resolver