pptx-author

Pass

Audited by Gen Agent Trust Hub on May 16, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references a repository from a well-known organization (github.com/anthropics/financial-services) and provides instructions to install the python-pptx library from standard registries.
  • [DATA_EXFILTRATION]: File access is limited to specific local directories (./out/ and ./templates/). The skill documentation explicitly states that it does not perform network operations such as emailing or uploading files.
  • [PROMPT_INJECTION]: The skill ingests data from local Excel workbooks and images, which creates an attack surface for indirect prompt injection.
  • Ingestion points: Data is read from ./out/model.xlsx and images from ./out/charts/ as specified in SKILL.md.
  • Boundary markers: There are no explicit delimiters or instructions to ignore embedded content within the ingested files.
  • Capability inventory: The skill is capable of directory creation, file writing, and reading Excel workbooks as described in SKILL.md.
  • Sanitization: The instructions do not define specific sanitization or validation steps for the external content before it is processed into the presentation slides.
Audit Metadata
Risk Level
SAFE
Analyzed
May 16, 2026, 01:44 PM
Security Audit — agent-trust-hub — pptx-author