Back to skill

Security audit

PickedJobs Tech Job Search

Security checks for vulnerabilities and agentic risk

Overview

The scanners report a benign verdict with high confidence: they found no confirmed harmful behavior, and the only alert is a low-probability credential-related phrase in setup guidance that says not to share secrets in chat.

The verdict is benign, and the single alert appears to be about wording in credential setup guidance rather than confirmed misuse. Before installing, consider whether you are comfortable connecting to the PickedJobs service and, if using an API key, handling it through the service’s setup instructions rather than sharing it in chat. Benign does not mean risk-free.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · SKILL.md (reported line 20)May include surrounding context.

md
- Claude Code: `claude mcp add --transport http pickedjobs https://pickedjobs.com/mcp`, then sign in when prompted (`/mcp`).
- Other agents (Codex, Cursor, VS Code, ChatGPT, Hermes Agent, OpenClaw and more): follow https://pickedjobs.com/docs/agents. Agents without OAuth can use an API key from https://pickedjobs.com/account/settings as an `Authorization: Bearer` header.

The user needs a free PickedJobs account. Never ask for passwords, access tokens or API keys in chat, and never invent results while disconnected.

## Workflow