able-harness-kit

v0.2.0 suspicious
4.0
Medium Risk

Thin, backend-neutral agent-harness middlewares for LangChain / deepagents: loop guard, binary-read guard, tool-result budget.

🤖 AI Analysis

Final verdict: SUSPICIOUS

The package shows minimal risks in terms of network usage, shell execution, obfuscation, and credential harvesting. However, its metadata raises concerns due to the repository's recent creation and limited contribution history.

  • Recent repository creation
  • Single contributor with few commits
Per-check LLM notes
  • Network: No network calls detected, which is normal unless the package requires internet access to function properly.
  • Shell: No shell execution patterns detected, indicating no direct command execution from the package.
  • Obfuscation: No obfuscation patterns detected, indicating low risk.
  • Credentials: No credential harvesting patterns detected, indicating low risk.
  • Metadata: The repository's recent creation, single contributor with few commits, and the author's lack of history suggest potential risk.

🔬 Heuristic Checks

✓ Outbound Network Calls

No suspicious network call patterns found

✓ Code Obfuscation

No obfuscation patterns detected

✓ Shell / Subprocess Execution

No shell execution patterns detected

✓ Credential Harvesting

No credential harvesting patterns detected

✓ Typosquatting

No typosquatting candidates detected

⚠ Registered Email Domain score 3.0

Suspicious email domain flags: Very short email domain: qq.com>

  • Very short email domain: qq.com>
✓ Suspicious Page Links

All external links appear legitimate

⚠ Git Repository History score 10.0

Git history flags: Repository created very recently: 5 day(s) ago (2026-05-31T13:10:29Z)

  • Repository created very recently: 5 day(s) ago (2026-05-31T13:10:29Z)
  • Repository has zero stars and zero forks
  • Single contributor with only 3 commit(s) — possibly throwaway account
  • All 3 commits happened within 24 hours
⚠ Maintainer History score 4.0

2 maintainer concern(s) found

  • Author name is missing or very short
  • Author "" appears to have only 1 package on PyPI (new or inactive account)
✓ Known CVE Vulnerabilities

No known vulnerabilities found in OSV database.

💡 AI App Starter Prompt

Use this prompt to build a project with able-harness-kit 
Create a fully functional mini-application named 'AgentGuard' using the Python package 'able-harness-kit'. This application will serve as a robust monitoring tool for agents running in a backend-neutral environment, ensuring they operate within safe parameters. The core functionalities of 'AgentGuard' include loop detection and prevention, safeguarding against unauthorized binary reads, and managing the output size of tool results to prevent excessive resource usage.

Step 1: Initialize your Python environment and install the 'able-harness-kit' package. Make sure to also include any necessary dependencies.

Step 2: Design and implement a basic agent framework within 'AgentGuard' that can be extended to run various types of agents. Ensure that this framework includes methods for starting, stopping, and querying the state of the agents.

Step 3: Utilize the 'loop guard' feature from 'able-harness-kit' to monitor agent execution. Implement a mechanism to detect and terminate infinite loops within the agents to ensure they do not consume unnecessary resources.

Step 4: Integrate the 'binary-read guard' functionality to protect against potential security risks associated with unauthorized file access. Agents should not be allowed to read arbitrary binary files unless explicitly permitted.

Step 5: Implement a tool result budget system using 'able-harness-kit' to manage the output size of tools executed by the agents. This ensures that no single tool's output becomes excessively large, which could overwhelm the system.

Suggested Features:
- User-friendly interface for adding, removing, and configuring agents.
- Detailed logs and reports for each agent's execution, including loop detection events and binary read attempts.
- Configurable thresholds for loop detection sensitivity and tool result size limits.
- Real-time monitoring dashboard to visualize agent performance and health status.

The goal is to create a versatile and secure platform for managing agents, leveraging the powerful capabilities provided by 'able-harness-kit'.