The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill provides Python examples for interacting with the Monday.com API that construct GraphQL queries using string interpolation (%s) for variables such as sprint_name, group_name, item_name, and column_id. If the agent processes data from a board containing malicious GraphQL control characters (e.g., in a board name or item name) and uses these snippets, it could lead to malformed queries or unintended API operations.
Ingestion points: Data ingested from Monday.com boards via search_items (items and column values) and verify_monday_access (account and product metadata) in SKILL.md and SKILL_HE.md.
Boundary markers: None. The instructions do not define delimiters or provide warnings for the agent to treat data retrieved from the API as untrusted or needing sanitization.
Capability inventory: The skill has significant capabilities to modify account data using the requests.post method for GraphQL mutations (creating boards/groups, updating column values, moving items) and has access to the Bash and WebFetch tools.
Sanitization: Absent. The code snippets rely on Python's string formatting operator rather than using GraphQL variables, which is the standard practice for preventing injection attacks.

monday-com-workflows