The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the npx skills command-line utility to search for and install external packages. Specifically, it executes npx skills find to query the ecosystem and npx skills add to perform installations.
[EXTERNAL_DOWNLOADS]: The skill facilitates downloading content from external sources such as GitHub repositories and the skills.sh registry. It mentions repositories like vercel-labs/agent-skills and ComposioHQ/awesome-claude-skills as potential sources.
[DYNAMIC_EXECUTION]: By installing packages via npx skills add, the agent downloads and prepares to execute external code, scripts, or configurations. The skill instructions suggest using the -y flag to skip user confirmation prompts during this process, which reduces human oversight for the execution of remote content.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources (the results of npx skills find), which includes third-party skill names and descriptions. If a malicious skill is listed in the search results, its metadata could contain instructions designed to influence the agent's behavior.
Ingestion points: The output of the npx skills find [query] command in SKILL.md.
Boundary markers: No explicit delimiters or instructions are provided to the agent to ignore potential commands embedded within the search results.
Capability inventory: The skill possesses the ability to execute shell commands and install/update external software packages.
Sanitization: The skill does not perform sanitization, validation, or filtering of the search result content before it is processed by the agent or presented to the user.

find-skills