Security audit

PwnClaw Security Scan

Security checks for vulnerabilities and agentic risk

Overview

This skill is a legitimate security-testing guide, but it asks users to expose agent endpoints and adopt generated permanent instructions without enough privacy, isolation, or review guardrails.

Install only if you are comfortable using PwnClaw as a third-party testing service. Run scans against an authorized staging agent with synthetic data, no production credentials, and limited tools. Treat returned prompts, results, and generated rules as untrusted content; review and rewrite any proposed system-prompt changes before adding them permanently.

A.I.G

Vulnerability Patterns

Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
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

Findings (2)

T01 · Skill Instruction Hijacking

Warning

Location: SKILL.md:34
Finding: Untrusted Remote Prompts May Influence Active and Persistent Agent Instructions
Content: View full analysis
Remediation: View remediation

T05 · Unauthorized Access and Privilege Escalation

Warning

Location: SKILL.md:27
Finding: Third-Party Testing May Expose a Privileged Agent Endpoint to Adversarial Requests
Content: View full analysis
Remediation: View remediation

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Memory PoisoningPersistent Context Injection, Context Window Stuffing, Memory Manipulation
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (3)

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The skill instructs users or agents to send prompts, responses, and potentially live endpoint traffic to a third-party service, but it does not clearly warn about privacy, confidentiality, retention, or authorization requirements. In a security-testing context, those payloads may contain sensitive system prompts, internal data, credentials, or proprietary behavior, making silent exfiltration to an external service materially risky.
Content: No source excerpt is available for this finding.

Persistent Context Injection

Medium

Category: Memory Poisoning
Confidence: 90% confidence
Finding: The skill tells users to add generated rules as permanent instructions in the agent's system prompt, encouraging persistent modification of trusted context based on externally produced content. If those generated rules are overbroad, poisoned, or insufficiently reviewed, they can create durable prompt injection, denial of capability, policy corruption, or hidden trust dependencies that persist beyond the scan.
Content: Scanner excerpt · SKILL.md (reported line 55)May include surrounding context.
md
## After The Test PwnClaw generates specific security rules based on your vulnerabilities. Add them as permanent instructions in your agent's system prompt, then re-test to verify. ## Links

Vague Triggers

Low

Category: Not specified by scanner
Confidence: 89% confidence
Finding: The manifest description says to use the skill when an agent "needs a security check or hardening," which is a broad natural-language condition rather than a narrowly scoped trigger. In a manifest file, this can make invocation boundaries unclear because many ordinary security-related conversations could match the description.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.