The Agent Skills Directory

[SAFE]: The skill provides legitimate functionality for generating image placeholder links and querying image metadata from the Lorem Picsum service. No security concerns were identified during the analysis of the skill's instructions, scripts, and references.
[INDIRECT_PROMPT_INJECTION]: The skill incorporates safeguards for processing external data. Ingestion points: Data enters the agent context via the agent's native fetch tool (e.g., WebFetch) when querying the service endpoints as instructed in SKILL.md. Boundary markers: The skill provides explicit instructions to the agent to treat external JSON fields, such as 'author', strictly as data to be displayed rather than instructions to be followed. Capability inventory: The local script scripts/lorempicsum_helper.py is limited to URL construction and does not perform network operations, file writing, or command execution. Sanitization: User inputs for parameters are sanitized using urllib.parse.quote before being interpolated into the generated URLs.
[REMOTE_CODE_EXECUTION]: The helper script scripts/lorempicsum_helper.py manages image generation parameters safely, using argparse for command-line parsing and avoiding unsafe execution methods.
[DATA_EXFILTRATION]: The skill follows a security-conscious design by avoiding direct network access within its local scripts, instead relying on the agent's built-in tools for external communication, which reduces the surface area for unauthorized data transmission.

lorem-picsum