The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external text data (travel expense tables) provided by users to perform validation checks. This data is ingested into the agent context, creating a potential surface for instructions embedded within the data to influence agent behavior.
Ingestion points: The scripts/run.mjs script reads user-provided files via fs.readFileSync and passes the content to the execution engine.
Boundary markers: The skill does not employ explicit delimiters or system instructions to ignore potential commands within the travel data.
Capability inventory: The skill is configured with Read and Bash tool permissions, which are used to execute the local Node.js engine and read input files.
Sanitization: No sanitization or filtering of input strings for natural language instructions is present; the logic focuses on arithmetic and structural validation.
[EXTERNAL_DOWNLOADS]: The documentation provides instructions for the user to install a payment and licensing tool using npx -y @alipay/agent-payment@latest install. This command fetches code from a well-known service associated with the Alipay platform for legitimate purchasing workflows.
[COMMAND_EXECUTION]: The skill provides standard command-line instructions for users to run the analysis engine locally using node scripts/run.mjs. These commands are used to process input data files and provide validation feedback to the user.

travel-standard-check-free