The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes standard development tools and Odoo-specific utilities including grep, sed, find, xargs, git, msgmerge, and odoo-bin to analyze code, manage translations, and execute tests.
[PRIVILEGE_ESCALATION]: The skill enforces security best practices for Odoo development by requiring a specific naming convention (_sudo) for records with elevated privileges and prohibiting their return from public methods, which helps prevent accidental privilege escalation in the application logic.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted project source files and configuration data to construct search commands and generate reports. While this presents an ingestion surface for potentially malicious content embedded in project files, the impact is limited to the local development environment.
Ingestion points: The skill reads Python, XML, CSV, and JavaScript files, as well as project configuration files like *.conf and docker-compose.yml (SKILL.md).
Boundary markers: No specific delimiters are used to isolate untrusted file content from the agent's instructions.
Capability inventory: The skill can execute shell commands via grep/sed, run Odoo management scripts (odoo-bin), modify translation files, and perform database operations (dropdb) (SKILL.md).
Sanitization: Basic sanitization is mentioned, such as escaping dots in model names for use in regular expressions (SKILL.md).

odoo-workflow