linkedin-profile-optimizer

Pass

Audited by Gen Agent Trust Hub on Apr 4, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection by processing untrusted user input from LinkedIn profiles during its audit and rewrite phases.\n
  • Ingestion points: The intake process defined in the "Step 1 — Intake" section of SKILL.md (and the "Required Inputs" section of SKILL-OC.md) requires the agent to collect and process a user's current headline, about section, and experience bullets.\n
  • Boundary markers: The skill lacks explicit boundary markers or delimiters (such as XML tags or triple quotes) and does not provide instructions to the agent to treat the user-provided text strictly as data.\n
  • Capability inventory: The agent uses the ingested profile data to perform scoring, generate rewritten headlines and about sections, and create social media posts.\n
  • Sanitization: No input validation, escaping, or filtering mechanisms are implemented to prevent embedded instructions within the user's profile text from influencing the agent's behavior.\n- [SAFE]: The skill includes a link to the author's library on GitHub (github.com/BrianRWagner/ai-marketing-skills). As a reference to a well-known service, this is documented neutrally and does not affect the security verdict.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 4, 2026, 09:05 AM