ClawHub
Back to skill

Security audit

TASD

Security checks for vulnerabilities and agentic risk

Overview

This is a Markdown-only design guide for agent-native services, with no executable code or hidden privileged behavior in the submitted artifacts.

This appears safe to install as a design aid. When using it to create real service specs, review any generated curl install commands, API-key examples, credential storage paths, heartbeat behavior, and agent write actions before publishing or letting agents act autonomously.

SkillSpector

Vulnerability Patterns
  • Rogue AgentSelf-Modification, Session Persistence
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Vague Triggers

Medium
Confidence
91% confidence
Finding
The manifest description is overly broad and can cause this skill to be invoked for many generic product-design requests, increasing the chance that unrelated or less-safe tasks are routed through agent-native guidance. That broad routing can expose users and downstream agents to unnecessary auth, automation, and autonomous-operation patterns in contexts where they are not appropriate.

Session Persistence

Medium
Category
Rogue Agent
Content
2. **Skill files table** — list every file in the bundle with its URL. Agents load additional files on demand.
3. **Install block** — copy-pasteable `curl` commands to install the full bundle locally. Also note that agents can read files directly from URLs without installing.
4. **Security contract** — a `CRITICAL SECURITY WARNING` block near the top. List the exact domain the API key must never leave. Use NEVER/REFUSE language. The agent must see this before it sees the API docs.
5. **Registration / first-run** — the minimal API call to create an identity. Show the full response shape including where the API key lives, what the `claim_url` is for, and what the human must do.
6. **Credential storage** — show the agent exactly where and how to persist its key (`~/.config/<service>/credentials.json` or env var).
7. **Auth pattern** — one representative `curl` with `Authorization: Bearer` header.
8. **Capability summary table** — every action the agent can take, one line each, with a priority signal (`🔴 Do first`, `🟠 High`, `🟡 Medium`, `🔵 When ready`).
Confidence
86% confidence
Finding
The skill instructs authors to return full response shapes including where the API key lives and to persist credentials in local files or environment variables, which normalizes long-lived secret handling by agents. In an agent ecosystem, this increases the risk of credential leakage through logs, prompt injection, tool misuse, overbroad file access, or accidental inclusion in generated documentation.

Static analysis

No suspicious patterns detected.