phab
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of data from a Phabricator instance, creating a surface for indirect prompt injection attacks where malicious instructions embedded in tasks or comments could influence agent behavior.
- Ingestion points: Several commands, including
phab task view,phab task search,phab revision search, andphab revision commentsinSKILL.md, retrieve external content from a Phabricator server into the agent's context. - Boundary markers: The skill lacks explicit instructions or boundary markers to treat retrieved Phabricator content as untrusted data or to ignore embedded instructions.
- Capability inventory: The skill utilizes the
phabCLI tool to execute various commands, including task creation, modification, and revision management, which are performed based on the context of the ingested data. - Sanitization: There is no evidence of sanitization, filtering, or validation of the retrieved text before it is presented to the agent's reasoning engine.
Audit Metadata