The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the agent to use npx skills add to fetch packages from remote sources, including GitHub repositories and the skills.sh registry. While it highlights some well-known repositories like vercel-labs/agent-skills, the tool is designed to pull content from any user-specified repository.
[COMMAND_EXECUTION]: The skill core functionality relies on executing shell commands through the npx skills CLI to manage, search, and install software on the host system.
[REMOTE_CODE_EXECUTION]: By design, this skill facilitates the execution of remote code by installing third-party agent skills. The documentation specifically encourages the use of the -y flag, which bypasses confirmation prompts, potentially leading to the silent installation of malicious extensions if the source repository is compromised or spoofed.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8).
Ingestion points: The agent ingests and processes untrusted data from the output of the npx skills find command, which includes package names and descriptions created by third parties.
Boundary markers: No explicit delimiters or instructions are provided to the agent to treat search results as untrusted data.
Capability inventory: The skill has the capability to execute shell commands (npx) and modify the agent's runtime environment by adding new skills.
Sanitization: There is no evidence of sanitization or validation of the search results before they are presented to the user or used for installation.

find-skills