agentkeeper-ai

v1.1.3 suspicious
4.0
Medium Risk

Cognitive continuity infrastructure for long-lived AI agents — cross-model state reconstruction, semantic recall, cognitive compression

🤖 AI Analysis

Final verdict: SUSPICIOUS

The package shows a moderate network risk and concerns over metadata, suggesting potential unauthorized data transmission and lack of maintainer transparency.

  • Moderate network risk due to external API communication
  • Metadata risk due to unclear maintainer information
Per-check LLM notes
  • Network: The observed network call pattern suggests the package may be communicating with an external API, which could indicate legitimate functionality but also potential for unauthorized data transmission.
  • Shell: No shell execution patterns were detected, indicating a low risk of immediate system compromise through shell command execution.
  • Obfuscation: No obfuscation patterns detected, indicating low risk.
  • Credentials: No credential harvesting patterns detected, indicating low risk.
  • Metadata: The maintainer has a new or inactive account and lacks detailed author information, raising some concerns but not definitive indicators of malice.

🔬 Heuristic Checks

⚠ Outbound Network Calls score 3.0

Found 2 network call pattern(s)

  • ).encode() req = urllib.request.Request( f"{self.host}/api/chat", da
  • son"}, ) with urllib.request.urlopen(req, timeout=self.timeout) as resp: body
✓ 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

Email domain looks legitimate: thinklanceai.com>

✓ Suspicious Page Links

All external links appear legitimate

✓ Git Repository History

Repository Thinklanceai/agentkeeper appears legitimate

⚠ 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 agentkeeper-ai 
Develop a personalized virtual assistant named 'Cognitive Companion' that leverages the 'agentkeeper-ai' library to provide users with a seamless and context-aware experience. This assistant will be capable of maintaining user-specific information across different sessions and interactions, ensuring that it can recall past conversations and tasks, thereby enhancing its responsiveness and relevance.

### Project Goals:
1. **User Authentication & Session Management:** Implement a system where users can sign up, log in, and maintain their session states using cookies or tokens.
2. **Cross-Session Continuity:** Utilize 'agentkeeper-ai' to store and retrieve user interaction history and preferences across different sessions, ensuring that the assistant remembers previous interactions and maintains a consistent persona.
3. **Contextual Recall:** Enable the assistant to recall specific contexts from past interactions, such as task status updates, reminders, or personal preferences, to provide more relevant responses.
4. **Semantic Compression:** Apply semantic compression techniques provided by 'agentkeeper-ai' to reduce the storage footprint of user data while preserving essential information for recall.
5. **Interactive Dashboard:** Create a web-based dashboard where users can view their interaction history, manage preferences, and initiate new queries or tasks.

### Key Features:
- **Personalized Greeting & Recommendations:** Based on past interactions, the assistant greets users with personalized messages and offers tailored recommendations.
- **Task Management:** Users can create, track, and manage tasks through the assistant, which stores task details and progress across sessions.
- **Reminder System:** The assistant can set reminders based on user preferences and recall these reminders during future sessions.
- **Knowledge Base Integration:** Incorporate a basic knowledge base (e.g., FAQs, common questions) that the assistant can access and learn from, improving over time.
- **Feedback Loop:** Allow users to rate the assistant's performance and provide feedback, which helps in refining future interactions.

### Utilizing 'agentkeeper-ai':
- **State Reconstruction:** Use 'agentkeeper-ai' to reconstruct the state of the assistant at the start of each session, including user preferences and interaction history.
- **Semantic Recall:** Implement a feature where the assistant can recall specific pieces of information or previous interactions by leveraging semantic recall capabilities.
- **Cognitive Compression:** Apply cognitive compression methods to efficiently store user data, ensuring minimal storage space usage without losing critical interaction details.

### Expected Outcome:
By the end of this project, you will have developed a sophisticated yet user-friendly virtual assistant that demonstrates advanced cognitive continuity and contextual awareness, significantly enhancing the user experience.