honest-link-outreach

Warn

Audited by Gen Agent Trust Hub on Nov 12, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill directs the agent to fetch a remote file from https://linkdr.com/SKILL.md if the initial prospect list is small. This constitutes a download of external instructions into the agent's context.
  • [DYNAMIC_EXECUTION]: The procedure involves dynamic loading of instructions from a remote URL (https://linkdr.com/SKILL.md). If the external resource is compromised, it could allow for hijacking the agent's behavior.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from target landing pages to generate email drafts. 1. Ingestion points: External URLs and local README.md files. 2. Boundary markers: No delimiters are specified to isolate external content. 3. Capability inventory: The agent can write to local files (seo/outreach-batch-YYYY-MM-DD.md) and interact with external research tools. 4. Sanitization: No explicit sanitization of the content retrieved from target URLs is provided.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Nov 12, 2026, 03:55 AM