The Agent Skills Directory

[COMMAND_EXECUTION]: The skill integrates several local shell commands and scripts used for manifest generation, validation, and testing within the project environment. These include npm run manifest, bash scripts/eval-lint.sh, npm run validate, and node scripts/skill-behavior-eval.js. These are intended for local development workflows and project maintenance.
[DYNAMIC_EXECUTION]: A Python script snippet is included in references/description-recipe.md for validating skill frontmatter. The script reads a file path provided as an argument and utilizes yaml.safe_load to check specific metadata requirements such as the origin tag and the character length of the description field.
[INDIRECT_PROMPT_INJECTION]: The skill instructions direct the agent to read a local configuration file to adapt its interaction style.
Ingestion points: 02-DOCS/wiki/harness/user-profile.md (referenced in SKILL.md).
Boundary markers: No explicit delimiters or warnings are specified for the content of this file.
Capability inventory: The skill facilitates file authoring, manifest management via npm, and execution of local verification scripts.
Sanitization: There are no mentioned filters or validation steps for the data retrieved from the user profile wiki page.
[EXTERNAL_DOWNLOADS]: The skill references the command npx @ericrisco/rsc, which utilizes the NPM package runner to execute a tool from the official registry. As the package is part of the @ericrisco organization, which corresponds to the skill author, it is categorized as a vendor-owned resource used for catalog management.

author-skill