tracking-tasks
Pass
Audited by Gen Agent Trust Hub on Oct 30, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes system CLI tools such as
gh(GitHub),glab(GitLab), and Linear to synchronize task metadata and lifecycle states with remote platforms. These operations depend on pre-existing user authentication in the environment. - [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes task data from external sources, including remote issue trackers and local task manifests. This creates an attack surface where malicious instructions could be embedded in task descriptions, titles, or comments, potentially influencing the agent's behavior when processing or displaying task summaries.
- Ingestion points: Processes data from
.agents/task-tracker.md, the.agents/tasks/directory, and the output of the/split-taskscommand. - Boundary markers: No explicit delimiters or instructions are used to distinguish untrusted task content from system instructions.
- Capability inventory: The skill has permissions to write to the local filesystem under the
.agents/directory and execute repository CLI tools to create or update remote issues. - Sanitization: The instructions do not specify any validation or sanitization routines for the content retrieved from external trackers or generated manifests.
Audit Metadata