The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to process and modify Django project files, which constitute untrusted external data.\n
Ingestion points: Django source files including models, views, forms, and templates located in the project workspace.\n
Boundary markers: The instructions lack explicit delimiters or specific directives for the agent to distinguish between legitimate code and malicious instructions embedded within the processed codebases.\n
Capability inventory: The skill utilizes workspace-write permissions, enabling it to perform persistent file modifications based on its analysis.\n
Sanitization: There is no mention of sanitizing or validating the contents of the workspace files before processing.\n- [METADATA_POISONING]: The skill's metadata field specifies a non-existent AI model (gpt-5.6-sol). This inclusion of fictitious technical specifications is deceptive and may lead to misjudgment of the skill's actual capabilities or the environment it is intended for.

django-developer