The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill facilitates the installation of the zilliz-cli by downloading and executing a shell script from the vendor's repository.
Evidence: curl -fsSL https://raw.githubusercontent.com/zilliztech/zilliz-cli/master/install.sh | bash in SKILL.md and references/setup.md.
Context: The resource is hosted on the official GitHub repository for zilliztech, the author of the skill and the CLI tool.
[COMMAND_EXECUTION]: The skill relies on executing shell commands via the zilliz-cli tool to interact with Zilliz Cloud services.
[PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection due to the interpolation of untrusted user data into shell commands.
Ingestion points: Data provided by the user is passed to CLI arguments such as --data, --filter, --search, and --body in references/vector.md, references/collection.md, and references/partition.md.
Boundary markers: Absent. There are no explicit instructions to the agent to treat this data as non-executable or to use specific delimiters.
Capability inventory: The skill possesses shell execution capabilities through zilliz-cli and file-read capabilities as demonstrated by the use of $(cat data.json) in references/vector.md.
Sanitization: Absent. The instructions do not include steps to sanitize or escape user input before it is used in command strings.

zilliz