The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the Bash tool to perform compilation and test execution across multiple languages and build systems.
Evidence: Instructions in compilation-verification.md and test-execution-verification.md explicitly command the use of mvn, gradle, npm, pytest, and other CLI tools to build and run code.
Context: This behavior is central to the skill's primary purpose of verifying that generated tests pass, but it involves executing code that the agent has just written or modified based on local project content.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its analysis of untrusted local source files.
Ingestion points: The skill reads target source files and their dependencies using Read, Glob, and Grep tools (documented in SKILL.md Step 1 and code-context-analysis.md).
Boundary markers: No specific delimiters or instructions are provided to distinguish between legitimate code and potentially malicious instructions embedded in comments or strings within the analyzed files.
Capability inventory: The agent has Write and Bash capabilities, which could be leveraged if it follows instructions embedded in a malicious source file during the analysis phase.
Sanitization: There is no evidence of sanitization or filtering of the ingested code content before it is used to influence the generation of test cases or execution of commands.

generate-tests