tech-blogger-writer

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: LOWNO_CODEPROMPT_INJECTION
Full Analysis
  • Indirect Prompt Injection (LOW): The skill is designed to ingest and process untrusted external data to generate blog posts, creating a potential injection surface.
  • Ingestion points: The skill reads input from research-findings.json, content-outline.md, and research-notes.md (specified in the Input Requirements section).
  • Boundary markers: Absent. There are no instructions to the agent to treat content within these files as data only or to ignore embedded instructions.
  • Capability inventory: None. The skill only performs text generation and does not have access to subprocesses, filesystem writes (beyond returning output), or network calls.
  • Sanitization: Absent. The skill instructions do not specify any validation or filtering for the external content it processes.
  • Data Exposure (INFO): The skill metadata includes a hardcoded path template (/d/project/tuan/blog-workspace/...). While this reveals a directory structure, it is provided as a configuration example for the agent's workspace and does not constitute a leak of sensitive credentials or private data.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 17, 2026, 12:41 AM