The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data from CSV, XLSX, and TSV files as described in the workflow of SKILL.md. This introduces a vulnerability surface where malicious instructions embedded within the data could influence the agent's behavior.
Ingestion points: The agent reads user-provided data files in Step 1 of the workflow.
Boundary markers: No explicit instructions are provided to the agent to ignore or delimit instructions found within the data files.
Capability inventory: The skill has access to the Bash tool to execute Python scripts, which could be exploited if an injection occurs.
Sanitization: There is no evidence of data sanitization or validation of the content of the files before processing.
[DYNAMIC_EXECUTION]: The skill uses the Bash tool to run dynamically generated Python pandas scripts to perform data profiling and quality checks. While this is the intended functionality of the skill, generating code based on untrusted data column names or values can lead to command injection if not properly handled by the agent.
[COMMAND_EXECUTION]: The skill explicitly grants the agent the ability to execute shell commands via Bash for Python script execution. This powerful capability is used to process the untrusted data files mentioned above.

data-explorer