Back to skill

Security audit

File Organization Plan and Duplicate Report

Security checks for vulnerabilities and agentic risk

Overview

This skill creates file organization and duplicate reports for a folder you choose, with clear limits and no source-file changes.

Choose a specific folder and keep reports private because they reveal filenames and hashes. Use an approved local model if report details must stay on-device. Prefer a reviewed, pinned installer version over @latest. Duplicate matches still require human review before deleting anything.

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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • 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
Findings (4)

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 · tests/organize.test.mjs (reported line 19)May include surrounding context.

js
),'x');fs.writeFileSync(path.join(root,'b'),'y');await assert.rejects(plan(root,{maxFiles:1}),/limit/);});
test('byte limit fails before hashing an oversized input',async()=>{const {root}=fixture();fs.writeFileSync(path.join(root,'a'),'large');await assert.rejects(plan(root,{maxBytes:1}),/limit/);});
test('symlink or junction children are skipped and never traversed',async()=>{const {home,root}=fixture();const other=path.join(home,'outside');fs.mkdirSync(other);fs.writeFileSync(path.join(other,'secret.txt'),'outside');fs.symlinkSync(other,path.join(root,'link'),process.platform==='win32'?'junction':'dir');const result=await plan(root);assert.equal(result.files.length,0);assert.equal(result.skipped[0].reason,'symlink-or-junction');});
test('symlink or junction input roots are refused',()=>{const {home,root}=fixture();const link=path.join(home,'root-link');fs.symlinkSync(root,link,process.platform==='win32'?'junction':'dir');assert.throws(()=>checkedPath(link,'directory'),/Symlink/);});

MCP server referenced without pinned version: 'npx clawhub'.

Medium
Category
MCP Rug Pull
Confidence
94% confidence
Finding

The command uses npx clawhub@latest, which downloads and executes a mutable package version rather than a reviewed, fixed release. A compromised or malicious future release could execute code with the installing user's permissions, exceeding the organizer's read-only source-file workflow. The detector incorrectly labels this an MCP server reference, but the underlying unpinned executable dependency risk is valid; the manual installation context limits immediate exposure.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx clawhub'.

Medium
Category
MCP Rug Pull
Confidence
94% confidence
Finding

The documented installation command uses npx clawhub@latest, which can download and execute a changing npm package version with the installing user's permissions. A compromised future release could access local files or modify the workspace, regardless of the organizer's read-only design. The finding incorrectly describes ClawHub as an MCP server, but the underlying unpinned executable dependency risk is valid; this is optional installation guidance, not evidence of malicious behavior.

Content

No source excerpt is available for this finding.

Scope Creep

Low
Category
Excessive Agency
Confidence
70% confidence
Finding

Skill's behavior or capabilities extend beyond its stated purpose. Scope creep allows an agent to perform actions unrelated to its documented functionality, increasing the attack surface.

Content

Scanner excerpt · LICENSE (reported line 7)May include surrounding context.

text
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.