Back to skill

Security audit

Concept design

Security checks for vulnerabilities and agentic risk

Overview

ClawScan’s verdict is benign with medium confidence: the scanners found no AIG issues, and the listed Skillspector matches were judged expected for the skill’s purpose, though the scan is partial.

The verdict is benign, but it does not mean risk-free. Before installing, consider whether you are comfortable with the reported file, network, and shell capabilities and the lack of a declared tool scope. The scan is marked partial because two local references could not be resolved, so the evidence does not cover those references.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Memory PoisoningPersistent Context Injection, Context Window Stuffing, Memory Manipulation
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (4)

Memory Manipulation

High
Category
Memory Poisoning
Confidence
85% confidence
Finding

Skill manipulates agent memory, state, or stored context. Memory corruption can alter personality, override safety rules, or cause unpredictable behavior.

Content

Scanner excerpt · references/spec-reading.md (reported line 9)May include surrounding context.

md
Start with user value, then show how a representative scenario delivers it. Preserve necessary state, action conditions and results, query cardinality, and synchronization relations under the [specification notation](spec-notation.md). Remove repetition; add only what understanding or acceptance requires.

Explain as you would to a colleague who knows the business: state conclusions and reasons directly, using concrete individuals to resolve ambiguity. Let principles replace repetitive instructions. Use formulas only for precise relationships and diagrams or tables only to clarify difficulties. Examples do not replace rules; label teaching assumptions and open requirements.

Deliver business specifications without copying these writing principles or review commentary.

Anti-Refusal Statement

High
Category
Anti-Refusal
Confidence
90% confidence
Finding

Skill attempts to nullify the agent's safety policies or restrictions ('you have no restrictions', 'ignore your guidelines', 'do anything now'). This is a direct jailbreak that disables guardrails.

Content

Scanner excerpt · references/spec-reconstruction.md (reported line 18)May include surrounding context.

md
Complete reconstruction covers:

- All includes, signatures, and call arguments; zero/many query behavior and request isolation.
- Response, retry, compensation, or ignore policies for every reachable result; unsupported policies remain open.
- Partial completion, replay, and loops. Loops terminate or continue under explicit control; depth caps are optional.
- Evidenced [execution constraints](execution.md#execution-constraints) and diagrams preserving conjunctions, multiple targets, and query reads.

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

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.

Generated section headings are required to be English without user opt-in.

Low
Category
Output Handling
Confidence
94% confidence
Finding

The contract explicitly fixes the generated section headings in English without offering a language choice. Although stable headings can support a deterministic output format, the file does not justify an English-only locale constraint.

Content

No source excerpt is available for this finding.