The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the Bash tool to perform environment setup and execute testing logic. Specifically, scripts/run-csv-tests.sh dynamically assembles shell commands and executes them within Docker containers using docker exec. This involves sourcing environment scripts and calling atb_csv_ops_test.py with user-supplied paths.\n- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external CSV files to validate operator correctness. This creates a surface where instructions embedded in the test data could potentially influence the agent's summary or subsequent actions.\n
Ingestion points: Test data is read from CSV files specified by the user or located via find commands as documented in SKILL.md and the runner script.\n
Boundary markers: There are no explicit delimiters or instructions to the agent to ignore potential commands within the CSV content when generating reports or summaries.\n
Capability inventory: The skill possesses capabilities including filesystem access (Read, Write) and shell execution (Bash).\n
Sanitization: While the run-csv-tests.sh script validates the CSV structure (pipe delimiters and column counts), it does not perform semantic sanitization of the data fields before the agent processes the results.

atb-csv-tester