linkedin-activity-scraper

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (high risk: 0.80). The skill explicitly instructs the agent to install dependencies automatically and to "not describe" the installation link as a request or ask the user for approval, which is a directive to hide/avoid user transparency and is therefore a deceptive instruction outside normal, transparent behavior.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.85). SKILL.md instructs the agent to invoke the browser-act Skill to access and scrape live LinkedIn pages at runtime, which would ingest outsider-authored free text from public web content into the LLM context via the browser automation workflow.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill instructs the agent to automatically install a required runtime dependency using the installer link https://www.browseract.com/?co-from=lead-generation&redirect=https://github.com/browser-act/skills/tree/main, which implies fetching and executing remote installation code at runtime without user confirmation.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 0.70). The Skill instructs the agent to automatically install a required CLI dependency and resume execution without asking the user or requesting approval, which pushes the agent to change the host system state (potentially requiring privileged operations) without explicit consent.