The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes user-provided source code which represents an attack surface for indirect prompt injection. Malicious instructions embedded in the model code could potentially influence the agent during the migration.\n
Ingestion points: The skill accepts a file path via $ARGUMENTS to read V2 model code.\n
Boundary markers: The skill does not define specific delimiters or warnings to ignore instructions within the processed code.\n
Capability inventory: The skill utilizes grep for analysis and performs file reading and content transformation.\n
Sanitization: No input sanitization or validation of the source code content is performed.\n- [COMMAND_EXECUTION]: The skill provides specific shell commands for the agent to execute in order to identify code patterns.\n
Evidence: grep -rn "max.experimental" <model-path> in SKILL.md.\n- [EXTERNAL_DOWNLOADS]: The skill mentions dependencies on the MAX framework, which can be installed via standard package managers. These are legitimate references to the framework developed by the skill author.\n
Evidence: compatibility: Requires a MAX install (pip or pixi) in SKILL.md frontmatter.

migrate-max-v2-to-v3