Back to skill

Security audit

Fix llms.txt

Security checks for vulnerabilities and agentic risk

Overview

The verdict is benign with medium confidence: the scanners found no AIG issues, but raised purpose-fit concerns and could not fully inspect one referenced script.

The verdict is benign, not risk-free. Before installing, consider whether the reported context-pack and dependency-discovery capabilities fit your needs, and note that one script was only partially inspected, so the available 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
  • 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)

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 24)May include surrounding context.

md
- If there is none, generate one with the MCP `llms_txt` tool, the bundled helper `scripts/ink.mjs`, or by fetching `https://llms.ink/<url>`:

Skill declares no tool scope ('permissions' or 'allowed-tools') but code capabilities were detected: env.

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.

Dependency-manifest discovery and public registry lookups are unrelated to repairing the owner's published llms.txt.

Medium
Category
Excessive Agency
Confidence
98% confidence
Finding

The dependencyDocs capability reads local package manifests and resolves their dependencies through public package registries to third-party documentation or repository URLs (L256–L270). This is exposed through the deps command at L324, but discovering dependency documentation is not required to check and repair the owner's llms.txt and is not declared in the manifest. Registry requests also disclose the queried dependency names to those services.

Content

No source excerpt is available for this finding.

The CLI produces context-pack artifacts rather than implementing the declared llms.txt repair workflow.

Medium
Category
Excessive Agency
Confidence
96% confidence
Finding

The manifest describes checking a site's published llms.txt and writing a corrected or new file into the repository's public directory after approval. At L312–L327, the implemented commands instead read sources, build or refresh context packs, and discover dependencies; their file outputs are JSON or context/change Markdown at caller-selected paths, not a corrected llms.txt. This file provides no corresponding repair-and-publish operation.

Content

No source excerpt is available for this finding.