The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits a surface for Indirect Prompt Injection. It retrieves test case content (including titles, steps, and expectations) from the ZenTao API using the zentao-cli testcase get and zentao-cli testcase list commands. This external data is ingested into the agent's context without explicit boundary markers or sanitization instructions to distinguish it from system instructions.
Ingestion points: Found in references/zentao-testcase-get.md and references/zentao-testcase-list.md where test case details are fetched.
Boundary markers: None identified in the prompt instructions to isolate retrieved test case content.
Capability inventory: The skill has extensive capabilities via zentao-cli (create, update, delete, result recording) as documented in the references/ directory.
Sanitization: No sanitization or validation of the retrieved content is performed before processing.
[COMMAND_EXECUTION]: The skill relies on the execution of the zentao-cli system binary to perform all operations. The instructions guide the agent to construct and execute shell commands using this tool, including parameters for IDs, titles, and steps. Users should ensure the zentao-cli binary is from a trusted source.

zentao-testcase