linkedin-agent
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external configuration files to define the agent's writing voice, creating a potential surface for instructions embedded in that data to influence behavior.
- Ingestion points: The agent is instructed to read content from
.liftli/voice.mdand~/.liftli/voice.md(SKILL.md). - Boundary markers: No specific delimiters or safety instructions are provided to the agent for processing these external files.
- Capability inventory: The skill can route to multiple sub-skills and provides instructions for installing additional components.
- Sanitization: There is no evidence of sanitization or validation logic for the content of the voice configuration files.
- [COMMAND_EXECUTION]: The skill documentation includes instructions for installing additional components using
npx(npx skills add liftli-ai/linkedin-agent-skills). These commands originate from the vendor's repository and are standard for the skill's ecosystem, though they involve executing code from a remote source.
Audit Metadata