The Agent Skills Directory

[COMMAND_EXECUTION]: The interactive Bash scripts scripts/create_bug_report.sh and scripts/generate_test_cases.sh contain a command injection vulnerability. The scripts use eval "$var_name=\"$input\"" to assign user-provided input to internal variables. If the input contains shell metacharacters (e.g., $(command) or ; command ;), the shell will execute the embedded commands with the privileges of the user running the script.
[DYNAMIC_EXECUTION]: The skill employs dynamic variable assignment using the eval builtin within its provided shell scripts. This pattern is used to handle input for generating bug reports and test cases, which is inherently risky when the input source can be influenced by untrusted external data.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process external Figma design files to generate testing artifacts, creating an attack surface where malicious content in the designs could manipulate the agent's output or actions.
Ingestion points: Figma design files accessed via URLs as specified in SKILL.md and references/figma_validation.md.
Boundary markers: Absent. The instructions do not specify delimiters or warnings to ignore instructions embedded within design data.
Capability inventory: The skill has the capability to write files to the local filesystem and execute local shell scripts (scripts/create_bug_report.sh and scripts/generate_test_cases.sh) as described in SKILL.md.
Sanitization: Absent. There is no evidence of input validation or escaping for data extracted from external design files before it is processed or passed to the execution scripts.

qa-test-planner