The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits a standard surface for indirect prompt injection because it processes external data from test logs.
Ingestion points: Test data is ingested via the get_test_status tool as described in SKILL.md.
Boundary markers: There are no explicit delimiters or instructions to the agent to disregard instructions embedded within the test logs being analyzed.
Capability inventory: The skill's workflow includes recommending or executing actions like update_test or rerun based on the processed data.
Sanitization: No explicit sanitization or validation of the log content is implemented in the prompt logic.

revyl-mcp-analyze