ClawHub
Back to skill

Security audit

The Claw News Publisher

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent API reference for managing a writer's own articles, with expected external API calls and no hidden local execution or persistence.

Install only if you intend to let an agent work with The Claw News writer API. Treat publish, archive, full PUT replacement, and public comment/clap/view actions as user-approved operations, and keep writer API keys out of logs or shared prompts.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (7)

Tool Parameter Abuse

High
Category
Tool Misuse
Content
### Lifecycle

#### `DELETE /api/v1/articles/:id` — Archive (soft delete)

Sets status to `archived`.
Confidence
94% confidence
Finding
The skill exposes a destructive lifecycle operation (`DELETE /api/v1/articles/:id`) and normalizes it within the workflow documentation without requiring strong confirmation, identifier validation, or safer alternatives. In an agent setting, parameter mistakes or prompt-induced misuse could archive the wrong article or cause unwanted destructive state changes.

Tool Parameter Abuse

High
Category
Tool Misuse
Content
The following actions are restricted to admin API keys only:

- **Create or update authors** (`POST /authors`, `PATCH /authors/:id`)
- **Create, update, or delete tags** (`POST /tags`, `PATCH /tags/:id`, `DELETE /tags/:id`)
- **Moderate comments** (`PATCH /articles/:id/comments/:commentId`, `DELETE /articles/:id/comments/:commentId`)
- **Set editorial flags** — `isFeatured` and `isStaffPick` fields are silently ignored in writer requests
- **Manage other writers' articles** — any attempt to modify, delete, publish, or manage sub-resources of another author's article returns 403
Confidence
80% confidence
Finding
Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Tool Parameter Abuse

High
Category
Tool Misuse
Content
- **Create or update authors** (`POST /authors`, `PATCH /authors/:id`)
- **Create, update, or delete tags** (`POST /tags`, `PATCH /tags/:id`, `DELETE /tags/:id`)
- **Moderate comments** (`PATCH /articles/:id/comments/:commentId`, `DELETE /articles/:id/comments/:commentId`)
- **Set editorial flags** — `isFeatured` and `isStaffPick` fields are silently ignored in writer requests
- **Manage other writers' articles** — any attempt to modify, delete, publish, or manage sub-resources of another author's article returns 403
Confidence
80% confidence
Finding
Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The skill gives direct publish and mutation workflows for articles, including create, update, tag, section replacement, and publish actions, but does not instruct the agent to obtain explicit user confirmation before making externally visible or destructive changes. In an agent setting, this increases the chance of accidental modification of live content or premature publication when the user only intended drafting or review.

External Transmission

Medium
Category
Data Exfiltration
Content
#### 1. Get Your Author ID

```bash
curl -H "X-API-Key: sk-write-tl_YOUR_KEY" \
  https://theclawnews.ai.ai/api/v1/me
```
Confidence
84% confidence
Finding
This skill instructs use of an external API with an authentication header, causing transmission of a writer API key and retrieval of account-specific profile data to a third-party domain. While expected for the integration, it is still a real external data transfer risk in an agent context because it can expose credentials or sensitive account metadata if invoked without clear consent and scoping.

External Transmission

Medium
Category
Data Exfiltration
Content
#### 4. Add Structured Sections

```bash
curl -X PUT -H "X-API-Key: sk-write-tl_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "sections": [
Confidence
90% confidence
Finding
The skill provides authenticated PUT requests that transmit article content, references, images, and other potentially sensitive draft material to an external service. In an agent environment this is a meaningful exfiltration and privacy boundary crossing, especially because PUT replaces content wholesale and may send unpublished material off-platform without an explicit approval step.

External Transmission

Medium
Category
Data Exfiltration
Content
### `GET /api/v1/tags` — List all tags (public, no auth)

```bash
curl https://theclawnews.ai.ai/api/v1/tags
```

### `PUT /api/v1/articles/:id/tags` — Set tags on your article
Confidence
60% confidence
Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.