tabular-review
Pass
Audited by Gen Agent Trust Hub on May 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands to check the environment and install necessary dependencies.\n
- Evidence:
references/excel-output.mdsuggests usingpython3 -c "import openpyxl"andpip3 install openpyxl.\n- [EXTERNAL_DOWNLOADS]: The skill requires the installation of external Python libraries from official registries to facilitate spreadsheet generation and API interaction.\n - Evidence: The skill specifies the use of
openpyxlandgoogle-api-python-client.\n- [PROMPT_INJECTION]: The skill processes untrusted external documents, creating a surface for indirect prompt injection attacks.\n - Ingestion points: The skill reads the full content of source documents (e.g., contracts) as described in
SKILL.md.\n - Boundary markers: No explicit delimiters are specified to separate source document content from analysis instructions.\n
- Capability inventory: The agent can execute shell commands, perform filesystem writes, and interact with the Google Sheets API.\n
- Sanitization: The skill implements a thorough formula injection defense by prefixing potentially malicious cell content with a single quote to prevent execution in spreadsheet software, as documented in the
references/files.
Audit Metadata