The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses Azure DevOps MCP tools to perform state-changing operations on external project boards.
Evidence: Invokes wit_create_work_item, wit_update_work_item, wit_work_items_link, and wit_add_work_item_comment to build the work item hierarchy.
[DATA_EXFILTRATION]: The skill reads local project configuration and specification files to transmit their content to the Azure DevOps service.
Evidence: Reads documentation/workflow-config.md for organization and project settings, and parses tasks.md, requirements.md, and design.md to populate work item fields.
[PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection by processing external data that influences its tool usage.
Ingestion points: Reads task and requirement data from documentation/specs/{feature}/tasks.md and requirements.md (Step 2).
Boundary markers: No explicit delimiters or instructions are used to isolate the parsed markdown content from the agent's logic.
Capability inventory: The skill has permissions to create, update, and comment on work items, as well as modify the local ticket-map.md and the remote Wiki.
Sanitization: No validation or sanitization of the input markdown text is performed before it is used to populate work item titles and descriptions.

ado-backlog-manager