xlsx

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS] (SAFE): The skill is sourced from 'anthropics/skills', which is a verified and trusted repository.
  • [DATA_EXFILTRATION] (SAFE): Operations are limited to local file reading and writing of spreadsheet formats (.xlsx, .csv). No network activity, non-whitelisted domain requests, or sensitive system file access (e.g., SSH keys, AWS credentials) were detected.
  • [COMMAND_EXECUTION] (SAFE): The code snippets use standard library functions for data manipulation. There are no instances of arbitrary command execution, shell spawning (subprocess.run), or privilege escalation attempts.
  • [INDIRECT_PROMPT_INJECTION] (LOW): This skill defines an attack surface by ingesting untrusted data from external spreadsheet files.
  • Ingestion points: pd.read_excel() and load_workbook() in SKILL.md.
  • Boundary markers: Absent; the skill does not explicitly instruct the agent to ignore instructions embedded within the spreadsheet data.
  • Capability inventory: Limited to local file read/write. No network access or dynamic execution (eval/exec) capabilities are present in the provided scripts.
  • Sanitization: Not present in the snippets; the skill relies on the underlying libraries (pandas/openpyxl) to handle file parsing.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 04:59 PM