The Agent Skills Directory

[OBFUSCATION]: Detected the use of hidden SOH (Start of Header, \u0001) control characters within shell command strings in SKILL.md. These characters are used in place of spaces (e.g., qa_agent.py"\x01assert-script-implementation), which is a common technique to evade detection by security scanners or log analyzers that look for specific command patterns.- [DYNAMIC_EXECUTION]: The skill's primary function is the automated generation of executable scripts (Bash and Playwright/TypeScript) which are then executed by the agent. This includes the use of a custom CLI wrapper qa_agent.py to run these generated scripts with environment variables.- [INDIRECT_PROMPT_INJECTION]: The skill processes test-cases.json (external data) to generate the executable test tasks. This creates a vulnerability surface where malicious instructions embedded in the test case data could influence the generated code, leading to arbitrary code execution.
Ingestion points: test-cases.json (read via validate-cases and used in generate-spec-tasks).
Boundary markers: Absent; the skill does not define delimiters or warnings for the agent to ignore instructions within the JSON data.
Capability inventory: File system write (creating scripts), shell execution (running scripts and qa_agent.py), and network access (via curl in generated scripts).
Sanitization: No sanitization or validation logic is described for the content of the test cases before they are converted into commands.- [COMMAND_EXECUTION]: The skill performs broad searches across multiple hidden directories in the user's home folder (~/.claude/skills, ~/.agents/skills, ~/.codex/skills) to locate its installation path.

qa-test-script-generator