The Agent Skills Directory

[SAFE]: The skill interacts exclusively with Azure DevOps via its official REST API on dev.azure.com. This is a well-known service, and the data access is scoped to work items and team settings.
[SAFE]: The skill implements strong credential safety instructions. It requires the use of environment variables (AZURE_DEVOPS_PAT, AZURE_DEVOPS_ORG) for authentication and contains explicit directives forbidding the agent from asking the user to paste tokens into the chat or printing them in the output.
[SAFE]: The use of curl is restricted to API data retrieval and does not involve executing remote scripts or unverified binary downloads.
[INDIRECT_PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it processes work item titles and descriptions from an external repository.
Ingestion points: Azure DevOps Work Item fields (titles, states, types) fetched via API calls described in SKILL.md.
Boundary markers: The skill uses structured Markdown tables to present the data, which helps separate data from instructions.
Capability inventory: The skill uses curl for network requests to Azure DevOps and manages local temporary files (body.json) for API request payloads.
Sanitization: No explicit sanitization of external text is defined, though the structured output format provides a natural delimiter.

boards-team-active-work