The Agent Skills Directory

[DYNAMIC_EXECUTION]: The skill employs moda canvas edit and the <generate> markup tag to execute JavaScript for modifying canvas nodes. Documentation in references/edit-code.md indicates these environments are sandboxed but notes that the sandbox is not a hard security boundary, presenting a risk if untrusted data influences the executed logic.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from user inputs, external files (via moda file show), and web pages (via moda web read). This data informs the generation of markup and scripts, creating a surface for injection attacks. Ingestion points: User-supplied briefs and topics in SKILL.md; external URLs and file paths in references/multi-unit-workflow.md. Boundary markers: The instructions do not provide explicit delimiters or safety instructions to separate untrusted data from instructions during processing. Capability inventory: The skill uses Bash(moda:*) tools with capabilities for network access (moda media, moda web), file writing (moda export), and dynamic code execution. Sanitization: No sanitization or validation mechanisms are described for data ingested from external or user-provided sources.
[COMMAND_EXECUTION]: The skill's primary functionality is implemented through the execution of moda CLI commands via the Bash tool, as specified in the allowed-tools configuration.

moda-social-linkedin