LinkedIn DM

Security checks across malware telemetry and agentic risk

Overview

This skill automates sending LinkedIn messages from your logged-in account and includes workflow choices aimed at avoiding automation detection, so it needs review before use.

Install only if you are comfortable letting the agent operate your logged-in LinkedIn account, send reviewed messages to selected connections, and record outreach details. Pay special attention to the detection-avoidance language and keep volume low, recipient lists explicit, and logging destinations private.

VirusTotal

56/56 vendors flagged this skill as clean.

View on VirusTotal

Risk analysis

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

#
ASI02: Tool Misuse and Exploitation
High
What this means

Your LinkedIn account could send automated outreach in a way that may trigger platform restrictions or violate platform rules, even if the messages are user-approved.

Why it was flagged

The skill is not just automating LinkedIn messaging; it gives explicit workflow guidance to reduce automation detection, including for flagged accounts.

Skill content
Option A — Chrome Relay (recommended for flagged accounts) ... Navigate to feed before every new profile. Jumping directly between profiles is the #1 automation detection trigger.
Recommendation

Use only with a small, reviewed list, understand LinkedIn account-risk implications, and avoid using the detection-avoidance workflow for unsolicited or high-volume outreach.

#
ASI03: Identity and Privilege Abuse
Medium
What this means

The agent acts through your LinkedIn identity and can send direct messages that recipients will see as coming from you.

Why it was flagged

The skill requires access to the user's logged-in LinkedIn session so it can send messages as that user.

Skill content
Browser is open with LinkedIn logged in
Recommendation

Confirm every batch before sending, keep the recipient list narrow, and monitor the account during use.

#
ASI06: Memory and Context Poisoning
Medium
What this means

Personal contact details and outreach history may be stored outside LinkedIn in a sheet or local file.

Why it was flagged

The workflow persists contact details, message contents, hooks, and notes to Google Sheets, with a local JSON fallback if Sheets is unavailable.

Skill content
gog sheets append <SHEET_ID> "Outreach!A:L" ... "<Full Name>","<Role/Title>","<Company>","<LinkedIn URL>","<Hook Used>","<Opener Text>","<Pitch Text>" ... Notes field — include useful context: prior conversation history
Recommendation

Use a private sheet with limited sharing, avoid storing unnecessary personal notes, and delete the local progress file when it is no longer needed.

#
ASI05: Unexpected Code Execution
Low
What this means

If the page state is confusing, the fallback could click a send button based on DOM matching rather than a clearly selected UI reference.

Why it was flagged

The workflow includes an in-page JavaScript fallback to locate and click LinkedIn's Send button when the normal browser automation reference is unavailable.

Skill content
Or use the JS fallback if the Send button ref is not visible: () => { const btns = Array.from(document.querySelectorAll('button')); ... if (btn) { btn.click(); return 'clicked'; } }
Recommendation

Use the normal visible-button flow when possible and verify the correct compose window before allowing any fallback send action.