The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill utilizes the opensrc utility to fetch source code from well-known package registries (NPM, PyPI) and GitHub repositories (e.g., npx opensrc fetch zod) for the purpose of code inspection and reuse analysis.
[COMMAND_EXECUTION]: The skill's instructions involve the use of bash for repository searches (grep, find, ls) and npx for executing the opensrc inspection tool.
[REMOTE_CODE_EXECUTION]: The skill contains patterns for retrieving and potentially executing logic from remote packages via npx opensrc, which downloads source code to a global cache for agent consumption.
[PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection as it requires the agent to read and analyze content from the local codebase and external dependencies, which could contain adversarial instructions.
Ingestion points: Repository files accessed via grep and find; external source code fetched via opensrc (SKILL.md).
Boundary markers: Absent; the skill does not define specific delimiters for separating analyzed code from its own instructions.
Capability inventory: The skill has access to the Bash tool, enabling arbitrary command execution.
Sanitization: No sanitization or validation of the ingested code content is specified before analysis.

extend-before-create