The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructions direct the agent to utilize a domain-specific CLI tool, ha, for validating and installing preset packages (e.g., ha preset validate, ha preset install). This is a legitimate part of the developer workflow described.
[DYNAMIC_EXECUTION]: The skill defines a configuration schema for presets that includes script-based entrypoints (e.g., scripts/run.mjs). These scripts are intended to be executed by the Harness Anything platform runtime to perform preset-specific tasks.
[INDIRECT_PROMPT_INJECTION]: The skill handles the review and migration of existing preset manifests and documentation, which serves as an ingestion point for potentially untrusted data.
Ingestion points: The agent processes existing preset.json and PRESET.md files when performing review or migration tasks as described in SKILL.md.
Boundary markers: The instructions do not specify the use of delimiters or warnings to ignore instructions embedded within the processed preset data.
Capability inventory: The skill allows for local file creation, script generation, and execution of the ha CLI tool.
Sanitization: There are no explicit instructions for sanitizing or validating the content of the preset fields before they are processed or interpolated.

preset-creator