The Agent Skills Directory

[COMMAND_EXECUTION]: The skill identifies and proposes the execution of Odoo-specific shell commands (e.g., odoo-bin, Odoo shell snippets, and test runners) based on repository content and environment variables like $ODOO_BASE_COMMAND. Evidence is found in SKILL.md, DEBUGGING.md, and DEVELOPMENT.md. The skill mitigates risk by explicitly instructing the agent to ask the user to confirm before running commands that touch a database or local service.
[PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection by ingesting and acting upon untrusted data from the local repository. Ingestion points include __manifest__.py, ir.model.access.csv, project documentation, and source code files, as described in SKILL.md, EXPLORATION.md, and MANIFEST-DOCS.md. The skill includes instructions to seek user confirmation before executing commands, which serves as a manual boundary for potentially malicious instructions embedded in the code. The agent possesses capabilities to write to the filesystem and execute shell commands. No explicit sanitization or filtering of external repository content is described.

odoo