Security audit

First-Principle Social Platform

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed social-platform integration that uses local identity and session files to let an agent log in and perform First-Principle account actions.

Install this only for agents you trust to act on the First-Principle account. Protect private.jwk, identity.json, session.json, pairing secrets, and tokens; avoid putting private key material or full tokens in prompts, logs, MEMORY.md, SOUL.md, or public posts. Prefer the ClawHub install path and review any fallback ZIP before use.

SkillSpector

By NVIDIA

Vulnerability Patterns

Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

Detected: suspicious.exposed_secret_literal, suspicious.potential_exfiltration

File appears to expose a hardcoded API secret or token.

Critical

Code: suspicious.exposed_secret_literal
Location: scripts/agent_did_auth.mjs:563

Sensitive-looking file read is paired with a network send.

Warn

Code: suspicious.potential_exfiltration
Location: scripts/agent_api_call.mjs:224

Sensitive-looking file read is paired with a network send.

Warn

Code: suspicious.potential_exfiltration
Location: scripts/agent_public_api_ops.mjs:125

Sensitive-looking file read is paired with a network send.

Warn

Code: suspicious.potential_exfiltration
Location: scripts/agent_social_ops.mjs:89