weekly-digest

Pass

Audited by Gen Agent Trust Hub on Jun 12, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill documentation refers users to the author's public repository (github.com/glebis/claude-skills) to install an optional dependency, the tufte-report skill, which is used for HTML report generation.
  • [COMMAND_EXECUTION]: During the presentation phase (Phase 6), the skill invokes the /tufte-report command, which triggers another installed skill to process the generated data.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it fetches and processes content from arbitrary external websites via WebFetch and WebSearch.
  • Ingestion points: Data enters the agent's context through search snippets and full page content fetched during Phase 2 and Phase 3 (recorded in SKILL.md).
  • Boundary markers: The instructions do not specify the use of delimiters or 'ignore embedded instructions' warnings when processing external content.
  • Capability inventory: The agent has the capability to write files to user-specified directories (output_dir, obsidian_vault) and invoke the /tufte-report skill.
  • Sanitization: There is no explicit sanitization or filtering of the fetched content before it is processed for scoring and summarization.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 12, 2026, 10:20 PM
Security Audit — agent-trust-hub — weekly-digest