Security audit

Vibe Clawing

Security checks for vulnerabilities and agentic risk

Overview

This is a markdown-only workflow skill that encourages agent delegation and local progress tracking, with no hidden code, network access, or destructive behavior found.

Install only if you want a coaching skill that nudges agents toward more autonomous workflows. Review any proposed automation before using it on high-cost, irreversible, credentialed, customer-facing, financial, or production actions, and keep the ~/vibe-clawing/ tracking files free of secrets or sensitive business details.

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

Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
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

Findings (3)

Missing User Warnings

Medium

Confidence: 91% confidence
Finding: This guidance explicitly encourages reducing per-step human review and letting agents run workflows end-to-end, while only briefly mentioning guardrails elsewhere and not tying this escalation to concrete risk controls for sensitive actions, user data, or system integrity. In an agent skill promoting greater autonomy, that omission can normalize unsafe deployment patterns where operators over-delegate before establishing reliable authorization boundaries, monitoring, rollback, and data-safety checks.

Missing User Warnings

Low

Confidence: 94% confidence
Finding: The skill explicitly instructs the agent to create and maintain a persistent profile file in the user's home directory, but provides no warning, consent step, retention guidance, or sensitivity limits for stored content. This can lead to unintended local persistence of behavioral or preference data, which increases privacy risk and may surprise users who did not expect the skill to write durable state.

Missing User Warnings

Low

Confidence: 84% confidence
Finding: This markdown file includes a concrete instruction to create `~/vibe-clawing/loops.md` and populate it with operational notes, but it does not disclose that this stores potentially sensitive workflow information on disk. For markdown files, user-facing guidance that affects user data should include a warning when it may impact privacy or system state.

Static analysis

No suspicious patterns detected.