agloom

v0.1.93 suspicious
6.0
Medium Risk

Production agent framework on LangChain/LangGraph: nine execution patterns, persistent memory, skills, feedback, multi-level HITL, MCP, AGP protocol, runtime bridge, and observability hooks.

🤖 AI Analysis

Final verdict: SUSPICIOUS

The package exhibits significant obfuscation techniques and shell execution capabilities, raising concerns about its intended use and potential for abuse.

  • High obfuscation risk due to use of eval with restricted builtins.
  • Moderate shell risk from subprocess execution.
Per-check LLM notes
  • Network: The network calls indicate the package uses httpx for HTTP requests, which is common for packages that need to interact with web services.
  • Shell: Subprocess execution can be risky if not properly sanitized or controlled, as it might execute arbitrary commands. This suggests potential for misuse but doesn't necessarily indicate malicious intent.
  • Obfuscation: The use of eval with restricted builtins suggests an attempt to execute code in a restricted environment, which could be used for obfuscation or malicious purposes.
  • Credentials: No direct evidence of credential harvesting is present.

📦 Package Quality Overall: Medium (5.4/10)

◈ Medium Test Suite 6.0

Partial test coverage signals detected

  • Test runner config found: pyproject.toml
◈ Medium Documentation 7.0

Some documentation present

  • Documentation URL: "Documentation" -> https://agloom.readthedocs.io
  • Detailed PyPI description (6893 chars)
○ Low Contributing Guide 4.0

No contributing guide or governance files found

  • Development Status classifier >= Beta
◈ Medium Type Annotations 5.0

Partial type annotation coverage

  • 474 type-annotated function signatures detected in source
◈ Medium Multiple Contributors 5.0

Limited contributor diversity

  • 1 unique contributor(s) across 85 commits in HELLOMEDHIRA/agloom
  • Single author but highly active (85 commits)

🔬 Heuristic Checks

⚠ Outbound Network Calls score 4.5

Found 3 network call pattern(s)

  • " try: async with httpx.AsyncClient(timeout=30.0, follow_redirects=True) as client:
  • y: async with httpx.AsyncClient(timeout=30.0) as client: r = await clien
  • x self._client = httpx.AsyncClient( headers=self._headers, time
⚠ Code Obfuscation score 2.0

Found 1 obfuscation pattern(s)

  • " try: return str(eval(expr, {"__builtins__": {}}, {})) except Exception as exc
⚠ Shell / Subprocess Execution score 10.0

Found 6 shell execution pattern(s)

  • try: r = subprocess.run([cli, "--version"], capture_output=True, text=True, timeout=
  • try: r = subprocess.run([cli] + args, shell=False) except FileNotFoundError:
  • None try: proc = subprocess.run( # noqa: S603 [ rg,
  • try: proc = subprocess.run( # noqa: S603 argv, cwd=str
  • try: proc = subprocess.run( # noqa: S602 cmd, cwd=str(
  • try: proc = subprocess.Popen(cmd, **popen_kw) # noqa: S603 except OSError as exc
✓ Credential Harvesting

No credential harvesting patterns detected

✓ Typosquatting

No typosquatting candidates detected

✓ Registered Email Domain

No author email provided

✓ Suspicious Page Links

All external links appear legitimate

⚠ Git Repository History score 2.5

Git history flags: Repository has zero stars and zero forks

  • Repository has zero stars and zero forks
⚠ 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 agloom 
Create a conversational agent named 'ConvoCraft' using the 'agloom' Python package, designed to assist users in crafting personalized recipes based on their dietary preferences and available ingredients. This mini-app will leverage the advanced features of 'agloom' such as persistent memory, skills, feedback loops, and multi-level Human-in-the-Loop (HITL) mechanisms to enhance user interaction and satisfaction.

Step-by-step guide:
1. **Setup**: Install the 'agloom' package and set up a basic agent framework.
2. **Skill Development**: Develop skills for the agent to understand user dietary preferences, recognize available ingredients, suggest recipes, and provide nutritional information.
3. **Persistent Memory**: Implement persistent memory so the agent can remember past conversations and preferences from the same user, enhancing personalization.
4. **Feedback Mechanism**: Integrate a feedback loop where users can rate the recipes suggested by the agent, allowing for continuous improvement.
5. **Multi-Level HITL**: Design a multi-level HITL system where human intervention can refine suggestions when the agent's initial output does not meet expectations.
6. **Observability**: Use observability hooks provided by 'agloom' to monitor and log interactions, enabling better analysis and future enhancements.
7. **User Interface**: Develop a simple command-line interface for users to interact with the agent.

Features:
- Dietary preference recognition (vegetarian, vegan, gluten-free, etc.)
- Ingredient availability check
- Recipe suggestion based on user input and past data
- Nutritional facts provision for each recipe
- User feedback collection and analysis
- Persistent memory to store user preferences and previous interactions
- Multi-level human intervention for refining suggestions
- Detailed logs and analytics through observability hooks