Atlas / Skills / affaan-m / Agentshield

AgentshieldSAFE

skills/affaan-m/agentshield

AI agent security scanner. Detect vulnerabilities in agent configurations, MCP servers, and tool permissions. Available as CLI, GitHub Action, ECC plugin, and GitHub App integration. 🛡️

Verdict
SAFE
Grade
B
Trust score
89 /100
Version
—
Hosts
—
License
MIT
Stars
1,276
01

Overview

AI agent security scanner. Detect vulnerabilities in agent configurations, MCP servers, and tool permissions. Available as CLI, GitHub Action, ECC plugin, and GitHub App integration. 🛡️

Read from source at commit 7943d9f5ae37OBSERVED · 2026-10-12
02

What it tells the agent

The instruction file, verbatim from the audited commit — this is the text the model reads, and the surface the audit's instruction layer examines. Quoted here so you can judge it without cloning anything.

```markdown
# agentshield Development Patterns

> Auto-generated skill from repository analysis

## Overview
This skill teaches the core development patterns and conventions used in the `agentshield` TypeScript codebase. It covers file naming, import/export styles, commit message conventions, and testing patterns. This guide is designed to help contributors quickly align with the project's established practices and workflows.

## Coding Conventions

### File Naming
- Use **camelCase** for file names.
  - Example: `userAgent.ts`, `apiHandler.ts`

### Import Style
- Use **relative imports** for referencing local modules.
  - Example:
    ```typescript
    import { fetchData } from './apiHandler';
    ```

### Export Style
- Use **named exports** for all modules.
  - Example:
    ```typescript
    // In apiHandler.ts
    export function fetchData() { ... }
    ```

### Commit Messages
- Follow the **Conventional Commits** specification.
- Use the `feat` prefix for new features.
  - Example:  
    ```
    feat: add user agent parsing to middleware
    ```
- Average commit message length: ~59 characters.

## Workflows

### Feature Development
**Trigger:** When adding a new feature to the codebase  
**Command:** `/feature-development`

1. Create a new branch for your feature.
2. Implement the feature using camelCase file naming and relative imports.
3. Use named exports for any new modules.
4. Write or update corresponding tests (`*.test.*` files).
5. Commit your changes using the `feat` prefix and a descriptive message.
6. Open a pull request for review.

### Testing
**Trigger:** When verifying code changes or before merging  
**Command:** `/run-tests`

1. Identify or create test files matching the `*.test.*` pattern.
2. Run the test suite using the project's test runner (framework unknown; check project scripts).
3. Ensure all tests pass before merging or submitting a pull request.

## Testing Patterns

- Test files follow the `*.test.*` naming convention (e.g., `apiHandler.test.ts`).
- The testing framework is not specified; check for scripts or documentation in the repository.
- Place tests alongside the code they test or in a dedicated test directory.

**Example Test File:**
```typescript
// apiHandler.test.ts
import { fetchData } from './apiHandler';

describe('fetchData', () => {
  it('should return expected data', () => {
    // test implementation
  });
});
```

## Commands
| Command              | Purpose                                         |
|----------------------|-------------------------------------------------|
| /feature-development | Start the feature development workflow          |
| /run-tests           | Run the test suite for the codebase             |
```
03

Trust audit

SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.

LayerWhat it checksResult
L0Provenance & inventoryPASS
L1Static analysis of the codeNA
L2Instruction surface (what it tells the agent)PASS
L3Class-specific surfacePASS
L4Behavioural (sandbox)SKIPPED

What the source does

Filesystem
none-observed
Network
none-observed
Shell
none-observed
Dependencies
pinned
Secrets in source
none-found

Findings (1)

LOWInventory / provenance · skill.no_frontmatter · CWE-1104
SKILL.md:1
Why it matters. SKILL.md lacks name/description frontmatter

Gates applied: no_behavioural_pass.

Audited 2026-10-12 · audit v0.4.1 · source sha 7943d9f5ae37full audit observations/trust-audit/skill/affaan-m__agentshield.json · Report an issue / request a re-scan
04

Audit history

Every audit this skill has had.

DateSourceVerdictGradeScoreChange
2026-10-127943d9f5ae37SAFEB89first audit
05

Questions

What does the Agentshield skill do?

AI agent security scanner. Detect vulnerabilities in agent configurations, MCP servers, and tool permissions. Available as CLI, GitHub Action, ECC plugin, and GitHub App integration. 🛡️

Is Agentshield safe to install?

The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean skill reads B.

What can Agentshield access on my machine?

The audit observed no filesystem, network or shell use at all in its source.

How current is this page?

The grade is for one exact copy of the source (7943d9f5ae37), read on 2026-10-12. The repository is watched, and a new audit runs when it changes — this is the first audit.

Advertisement