ClawHub
Back to skill

Security audit

Spec Coder

Security checks across malware telemetry and agentic risk

Overview

This is an instruction-only spec-first coding workflow whose project file changes and review auto-approval behavior are disclosed and aligned with its purpose.

Install this if you want an assistant to manage a structured spec-first coding process in your project. Keep the repo under version control, review generated specs and code diffs, avoid putting secrets in spec/status files, and do not enable broad auto-approval preferences if you want explicit confirmation at every gate.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
Findings (4)

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
| **Medium** | Standard | 2–3 roles per gate. |
| **Large** | Full | All listed roles per gate. |

### Auto-Approve Rule

If the review finds **zero Critical and zero Major** issues, present a summary and auto-approve:
Confidence
91% confidence
Finding
The protocol allows the agent to proceed to the next phase automatically when no Critical or Major issues are found, which is an autonomous workflow decision with real downstream effects. In a spec-first coding skill, this can reduce user oversight and let flawed requirements, design, or code advance without explicit human approval, especially if the review itself misses security or safety issues.

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
### Auto-Approve Rule

If the review finds **zero Critical and zero Major** issues, present a summary and auto-approve:

> "Expert review complete — no Critical or Major issues found. [N Minor suggestions listed below.] Proceeding to next phase. Reply 'hold' to pause and address Minor items first."
Confidence
95% confidence
Finding
The quoted behavior explicitly says the workflow will proceed unless the user interrupts with 'hold', which makes continuation the default. That opt-out model is risky because users may miss the prompt, and the skill may continue making consequential development decisions without affirmative consent.

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
- `"Skip UX reviews"` — UX/UI Designer role is omitted from Gate 2.
- `"Focus on security"` — Security Expert role is always included, even on Small track.
- `"Auto-approve all gates"` — all gates auto-approve regardless of severity (user takes full responsibility).
- `"Only flag Critical"` — Major and Minor issues are listed but don't require user response.

If no preferences are set, use the default behavior (track-based depth, standard severity rules).
Confidence
97% confidence
Finding
The same preference setting creates a broad user-configurable bypass of review safeguards, and its placement in persistent review preferences increases the chance that unsafe behavior becomes the default for future runs. Because this is a development automation skill, persistent auto-approval can normalize skipping security and quality gates across multiple artifacts and iterations.

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
- `"Skip UX reviews"` — UX/UI Designer role is omitted from Gate 2.
- `"Focus on security"` — Security Expert role is always included, even on Small track.
- `"Auto-approve all gates"` — all gates auto-approve regardless of severity (user takes full responsibility).
- `"Only flag Critical"` — Major and Minor issues are listed but don't require user response.

If no preferences are set, use the default behavior (track-based depth, standard severity rules).
Confidence
97% confidence
Finding
The same preference setting creates a broad user-configurable bypass of review safeguards, and its placement in persistent review preferences increases the chance that unsafe behavior becomes the default for future runs. Because this is a development automation skill, persistent auto-approval can normalize skipping security and quality gates across multiple artifacts and iterations.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.