ClawHub
Back to skill

Security audit

Wiki Lint (Dexio)

Security checks for vulnerabilities and agentic risk

Overview

This is a transparent wiki-linting skill that reads a chosen markdown wiki, reports issues, and gives user-directed repair guidance.

Use it on wiki folders you intend to lint. It will read all markdown under the selected root and may report masked secret-like values with file and line information, so review output and any follow-up edits before committing or scheduling it.

A.I.G

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

SkillSpector

Vulnerability Patterns
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (4)

Tp4

High
Category
MCP Tool Poisoning
Confidence
93% confidence
Finding

The core health-checking behavior matches the description well: the script analyzes markdown wiki content for broken links, secrets, orphaned/unreferenced pages, frontmatter issues, stale pages, oversized pages, and duplicate titles, and it is standard-library-only. However, the description materially overstates capability by claiming it will 'fix what it finds.' The supplied code performs read-only analysis, reporting, and ranking; it never edits files or repairs links/frontmatter/content. That is a meaningful description-to-behavior mismatch. Additionally, the code optionally accesses git history via subprocess to compute better change times and verification ranking, which is an undeclared resource/capability, though secondary to the main mismatch.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
70% confidence
Finding

Without declared permissions the skill's intent is opaque and cannot be validated.

Content

No source excerpt is available for this finding.

subprocess module call

Medium
Category
Dangerous Code Execution
Confidence
70% confidence
Finding

subprocess module calls execute external commands. Without careful input validation, this enables command injection.

Content

Scanner excerpt · scripts/wiki_lint.py (reported line 283)May include surrounding context.

python
"""Last commit time per page when the wiki is a git checkout. File mtimes are
    reset by every clone, so they are a poor record of when a page changed."""
    try:
        top = subprocess.run(["git", "-C", str(root), "rev-parse", "--show-toplevel"],
                             capture_output=True, text=True, timeout=20)
        if top.returncode:
            return {}

subprocess module call

Medium
Category
Dangerous Code Execution
Confidence
70% confidence
Finding

subprocess module calls execute external commands. Without careful input validation, this enables command injection.

Content

Scanner excerpt · scripts/wiki_lint.py (reported line 287)May include surrounding context.

python
capture_output=True, text=True, timeout=20)
        if top.returncode:
            return {}
        log = subprocess.run(["git", "-C", str(root), "log", "--format=@%ct",
                              "--name-only", "--", "."],
                             capture_output=True, text=True, timeout=60)
    except (OSError, subprocess.SubprocessError):

Static analysis

No suspicious patterns detected.