backlog-md
Pass
Audited by Gen Agent Trust Hub on Jun 15, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill defines a set of interactions with the
backlogCLI. All commands are specific to project management (tasks, docs, search) and are executed locally within the workspace to manage task metadata and board state. - [DATA_EXPOSURE_AND_EXFILTRATION]: The skill attempts to derive a user handle from
git config user.name. This access is limited to a single non-sensitive configuration value and is used solely for local task assignment purposes within the CLI workflow. - [INDIRECT_PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it reads task and document content from local files.
- Ingestion points: Task and document data read via
backlog task <id>andbacklog doc view(SKILL.md, reference/cli-reference.md). - Boundary markers: None identified in the provided instructions.
- Capability inventory: Limited to
backlogCLI operations such as creating, editing, and listing tasks. - Sanitization: None explicitly mentioned; however, the requirement for agents to share implementation plans and wait for user approval provides a human-in-the-loop safeguard against unintended actions.
Audit Metadata