odoo-upgrade-planner
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill retrieves source code from well-known and community-trusted Odoo repositories on GitHub, specifically
odoo/odoo,odoo/enterprise, andOCA/OpenUpgrade. It also queries the official Odoo Apps Store atapps.odoo.comto identify potential module replacements. These actions are standard for the skill's purpose and utilize trusted sources within the Odoo ecosystem.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted Odoo module source code (Python, XML, SCSS, JS) which creates a standard indirect prompt injection attack surface.\n - Ingestion points: Data enters the agent's context when the skill reads and greps files within the user-provided module directory (e.g., manifest files, model definitions, and views).\n
- Boundary markers: The instructions lack specific delimiting markers or ignore-embedded-instruction warnings for the content extracted from the modules.\n
- Capability inventory: Across its scripts, the skill performs file reads, network lookups to verified Odoo and GitHub domains, and writes task files to the local file system.\n
- Sanitization: The skill does not implement explicit sanitization, validation, or escaping of the ingested source code content before it is used to generate tasks.\n
- Assessment: The risk is mitigated by the fact that the skill focuses on static analysis and report generation rather than dynamic code execution or administrative operations based on the untrusted input.
Audit Metadata