The Agent Skills Directory

[PROMPT_INJECTION]: The skill is potentially vulnerable to indirect prompt injection because it interpolates user-supplied data (such as search queries and destination names) into shell command arguments for the flyai-cli tool.
Ingestion points: User-provided search terms in fliggy-fast-search.md, search-flight.md, search-hotels.md, and search-poi.md (e.g., --query, --origin, --dest-name).
Boundary markers: Absent; there are no delimiters or specific instructions to the agent to ignore malicious commands embedded in the user's travel query.
Capability inventory: The skill performs shell command execution using the node runtime as specified in the SKILL.md metadata.
Sanitization: No explicit validation or escaping of input data is defined within the skill files before passing arguments to the CLI.
[COMMAND_EXECUTION]: The skill's primary functionality relies on executing the flyai-cli command-line utility. This involves subprocess calls that pass user-influenced parameters to the shell.
[EXTERNAL_DOWNLOADS]: The skill instructions specify the installation of the @fly-ai/flyai-cli package from the npm registry. This package originates from the author's official namespace and is consistent with the skill's stated purpose.

flyai