The Agent Skills Directory

[CREDENTIALS_UNSAFE] (HIGH): The file SKILL.md contains a hardcoded Google Gemini API key (AIzaSyDvvGGRbH4Os3Er0dYi0kE_AzE3_2b_Az8) in both the Phase 4 command example and the API Configuration table. This allows unauthorized access to the associated Google Cloud account.\n- [COMMAND_EXECUTION] (MEDIUM): The skill workflow in SKILL.md requires the agent to execute a local Python script scripts/generate_image.py using a shell command. This creates a potential command injection vector if the user-provided story description (interpolated into the --prompt argument) contains shell metacharacters and is not correctly escaped by the execution environment.\n- [DATA_EXFILTRATION] (LOW): The script scripts/generate_image.py makes outbound network requests to generativelanguage.googleapis.com using urllib.request.urlopen. Although this is functional for the skill's purpose, the domain is not in the trusted whitelist.\n- [PROMPT_INJECTION] (LOW): The skill is susceptible to indirect prompt injection as it processes untrusted user stories.\n
Ingestion points: User-provided story text in Phase 1 of SKILL.md.\n
Boundary markers: Absent; the instructions do not use delimiters to wrap the user input when constructing the image generation prompt.\n
Capability inventory: Subprocess execution and network requests via scripts/generate_image.py.\n
Sanitization: Absent; user input is directly interpolated into the prompt string without validation or escaping.

storyboard-generator