soia-pkm-clip-web

Pass

Audited by Gen Agent Trust Hub on Jul 22, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill provides installation instructions using npx skills add soia-team/soia-open-pkm-clip-skills, which downloads the skill package from the vendor's repository.
  • [COMMAND_EXECUTION]: To support PDF export on macOS, the skill uses osascript and System Events to perform UI automation within the Obsidian application.
  • [COMMAND_EXECUTION]: The skill utilizes several command-line utilities for PDF validation and processing, including pdfinfo, pdftoppm, and pdftotext as documented in references/obsidian-pdf-export.md.
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it ingests untrusted data from arbitrary web URLs.
  • Ingestion points: Article content is fetched from user-provided URLs using trafilatura or readability-lxml as specified in SKILL.md.
  • Boundary markers: The skill lacks explicit instructions or delimiters to ignore potential malicious instructions embedded within the scraped web content.
  • Capability inventory: The skill has the authority to write files to the local file system (Obsidian vault) and execute shell commands/UI automation scripts (osascript) as detailed in SKILL.md and references/obsidian-pdf-export.md.
  • Sanitization: There is no evidence of content sanitization or instruction filtering applied to the extracted web text before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 22, 2026, 01:46 PM
Security Audit — agent-trust-hub — soia-pkm-clip-web