Back to skill

Security audit

skills-best-practices

Security checks for vulnerabilities and agentic risk

Overview

This skill provides coherent guidance for writing skills, with a routine dependency risk in its suggested validation command.

Use a reviewed, pinned validator version in a limited project environment. The external validator was not executed or audited. Use authenticated troubleshooting for your own publications, and preserve accurate capability disclosures when applying the moderation wording advice.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (4)

External Script Fetching

High
Category
Supply Chain
Confidence
90% confidence
Finding

Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.

Content

Scanner excerpt · references/clawhub-publishing.md (reported line 29)May include surrounding context.

md
Only `suspicious.env_credential_access` is externally self-clearable; every other code requires a fixed re-publish.

**Hard-block codes** (`malicious.install_terminal_payload`, `malicious.crypto_mining`, `malicious.known_blocked_signature`) auto-hide the skill and place the uploader in manual moderation. The most common is `install_terminal_payload`: install instructions telling users to paste obfuscated shell payloads (base64-decoded `curl | bash`). Never include these, even as examples.

## Surviving the LLM Review

MCP server referenced without pinned version: 'uvx --from skills-ref'.

Medium
Category
MCP Rug Pull
Confidence
65% confidence
Finding

uvx/uv tool run commands without ==version create a rug-pull risk.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'uvx --from skills-ref'.

Medium
Category
MCP Rug Pull
Confidence
65% confidence
Finding

uvx/uv tool run commands without ==version create a rug-pull risk.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'uvx --from skills-ref'.

Medium
Category
MCP Rug Pull
Confidence
88% confidence
Finding

Although this is a validator rather than an MCP server, the skill explicitly instructs execution through uvx --from skills-ref without a version pin. That permits execution of a changing package release with the invoking environment's privileges; a compromised future release could access project files or credentials. Its placement in the normal pre-publication workflow makes execution plausible, though no malicious package or payload is demonstrated.

Content

No source excerpt is available for this finding.