Back to skill

Security audit

source-doc-slimming

Security checks for vulnerabilities and agentic risk

Overview

ClawScan marks this skill suspicious because it reports a mismatch between read-only permissions and instructions to modify files; the scan was partial, so evidence is incomplete.

Review what files the workflow would change and whether you intend to grant the required write access before installing. The scan was partial: although both listed components were scanned, several local references could not be resolved, so the evidence is incomplete.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
Findings (1)

The workflow requires modifying files despite declaring only file-read permission.

Medium
Category
Excessive Agency
Confidence
97% confidence
Finding

The skill declares only Read/file_read access but directs the agent to move content, rewrite source documents, create append helpers, and modify existing validation gates. If an agent treats those workflow instructions as authorization to use broader ambient capabilities, it could modify authoritative rules or configuration outside the declared permission scope; strict permission enforcement would instead block the workflow. These writes serve the skill's stated purpose and do not appear malicious, but its losslessness checks do not resolve the authorization mismatch.

Content

No source excerpt is available for this finding.