atlasagent-cli

v0.5.0 suspicious
6.0
Medium Risk

Atlas — production-grade anti-drift agent infrastructure. Ships seven cortex hooks that keep LLM agents from fabricating, drifting, or claiming work they have not actually shipped.

🤖 AI Analysis

Final verdict: SUSPICIOUS

The package exhibits multiple suspicious behaviors such as potential credential harvesting, obfuscated code, and questionable metadata, raising concerns about its legitimacy and potential for malicious activity.

  • credential risk of 8/10
  • obfuscation risk of 7/10
  • non-existent git repository
Per-check LLM notes
  • Network: The detected network calls seem to be for token exchange and retrieval, which could be legitimate if the package interacts with an API.
  • Shell: The shell execution patterns include commands that may be used for password management and git operations, suggesting possible interaction with system-level tools but requiring further investigation to confirm legitimacy.
  • Obfuscation: The presence of Base64 decoding suggests possible obfuscation, which could be used to hide code logic or evade simple static analysis.
  • Credentials: The detection of methods to read sensitive files and gather input securely indicates potential risks for harvesting credentials or other secrets.
  • Metadata: The package shows several red flags including a non-existent git repository, a single release from a potentially new or inactive author, and no maintainer history.

📦 Package Quality Overall: Low (4.6/10)

◈ 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://atlasagent.si/docs
  • Detailed PyPI description (11387 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

  • 720 type-annotated function signatures detected in source
○ Low Multiple Contributors 1.0

Could not retrieve contributor data from GitHub

  • GitHub API error: 404

🔬 Heuristic Checks

⚠ Outbound Network Calls score 9.0

Found 6 network call pattern(s)

  • oken_endpoints: req = urllib.request.Request( endpoint, data=data,
  • try: with urllib.request.urlopen(req, timeout=10) as resp: result = j
  • }).encode() req = urllib.request.Request( _OAUTH_TOKEN_URL, data=exch
  • OST", ) with urllib.request.urlopen(req, timeout=15) as resp: result = json.
  • encode("utf-8") request = urllib.request.Request( url, data=data, method="POST", head
  • , ) try: with urllib.request.urlopen(request, timeout=timeout) as response: r
⚠ Code Obfuscation score 10.0

Found 6 obfuscation pattern(s)

  • , 1)[0] raw = base64.b64decode(encoded) except Exception: conti
  • d>.<ext>``. """ raw = base64.b64decode(b64_data) ts = datetime.datetime.now().strftime("%Y%m%d_
  • [0]) try: __import__(import_name) except ImportError: missing.append(dep)
  • try: __import__("modal") except ImportError: print_info(
  • lled try: __import__("daytona") except ImportError: print_info("Install
  • '") try: __import__("vercel") except ImportError: print_info("Install
⚠ Shell / Subprocess Execution score 10.0

Found 6 shell execution pattern(s)

  • ssword entry result = subprocess.run( ["security", "find-generic-password",
  • can interact try: subprocess.run([claude_path, "setup-token"]) except (KeyboardInterrupt,
  • e]: try: result = subprocess.run( ["git", *args], cwd=cwd,
  • ne: try: result = subprocess.run( ["rg", "--files", str(path.relative_to(cwd))],
  • try: proc = subprocess.Popen( [self._acp_command] + self._acp_args,
  • nic() try: proc = subprocess.run( argv, input=stdin_json,
⚠ Credential Harvesting score 10.0

Found 4 credential access pattern(s)

  • "/etc/sudoers", "/etc/passwd", "/etc/shadow", ] } def build_wr
  • "/etc/passwd", "/etc/shadow", ] } def build_write_denied_prefixes(home: s
  • n.isatty(): val = getpass.getpass(prompt="") else: val = sys.stdin.readlin
  • getpass value = getpass.getpass(color(display, Colors.YELLOW)) else: val
✓ Typosquatting

No typosquatting candidates detected

✓ Registered Email Domain

Email domain looks legitimate: atlasagent.si>

✓ Suspicious Page Links

All external links appear legitimate

⚠ Git Repository History score 3.0

Repository not found (deleted or private)

  • Repository not found (deleted or private)
⚠ Maintainer History score 6.0

3 maintainer concern(s) found

  • Only one version has ever been released — brand new package
  • 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 atlasagent-cli 
Create a Python-based project named 'DriftGuard' which aims to monitor and ensure the integrity of AI responses generated by various language models in a production environment. This tool will utilize the 'atlasagent-cli' package to enforce strict adherence to the guidelines set forth by the developers, ensuring that the AI does not fabricate information, drift from its intended purpose, or claim to have completed tasks it has not actually performed.

Steps to build 'DriftGuard':
1. Set up a virtual environment and install the required packages, including 'atlasagent-cli'.
2. Define a configuration file where users can specify the acceptable behavior and boundaries for the AI responses.
3. Implement a monitoring system that checks each response against the specified rules using the hooks provided by 'atlasagent-cli'.
4. Develop a logging mechanism to record any instances where the AI deviates from the expected behavior.
5. Create a user-friendly interface that allows administrators to view logs, modify configurations, and receive alerts when potential issues arise.
6. Test 'DriftGuard' with different scenarios to ensure it effectively prevents drift and maintains the integrity of AI responses.

Suggested Features:
- Customizable rule sets for different use cases.
- Real-time alerting for detected anomalies.
- Historical data analysis to identify trends and patterns in AI behavior.
- Integration with popular chatbot frameworks for seamless deployment.

How 'atlasagent-cli' is Utilized:
- The 'atlasagent-cli' package will be used to integrate the necessary hooks into the application's workflow. These hooks will act as checkpoints, verifying that each response complies with the established rules before being delivered to the end-user. Additionally, 'atlasagent-cli' will provide the tools needed to log and analyze these interactions, ensuring that the AI remains within the defined parameters at all times.

💬 Discussion Feed

Leave a comment

No discussion yet. Be the first to share your thoughts!