Neron
v1.0.0Personal knowledge graph. Record notes, track moods, manage tasks, spot patterns in someone's life.
⭐ 0· 98·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
medium confidencePurpose & Capability
The skill is a connector for a personal knowledge graph and all described tools (search, semantic_search, cypher, create/update/delete, node_context, etc.) match that purpose. It does not declare unrelated binaries or environment variables. The MCP endpoint and token-based auth model described are coherent with a remote graph service.
Instruction Scope
SKILL.md and auxiliary docs explicitly instruct the agent to call a remote MCP endpoint and use rich tools (including raw Cypher). The docs also instruct users to obtain tokens/passwords via a Telegram bot and place tokens in agent config files. That behavior is within scope for an agent that should read/write a user's graph, but it grants powerful read/write access to sensitive personal data and instructs storing tokens in local config locations — the scope is broad by design and should be treated as high-sensitivity.
Install Mechanism
There is no install spec or executable code; the skill is instruction-only and does not download or install third-party binaries or archives. This minimizes local code-execution risk.
Credentials
No environment variables are declared in metadata, and the service uses per-user Bearer tokens obtained from a Telegram bot. Requiring a token for full read/write access to the user's graph is proportionate to the stated functionality, but those credentials are highly sensitive. The skill does not ask for unrelated credentials, but it does rely on the user placing tokens into agent config files (clear-text storage by instruction), which has privacy implications.
Persistence & Privilege
The skill is not marked always:true and does not request system-wide modifications. It instructs adding a connector/token to agent config (normal for connectors). It does not request elevated or persistent platform privileges beyond normal agent connectors.
Assessment
This skill is coherent with its purpose (a personal knowledge-graph connector) but it gives any connected agent full read/write access to very sensitive personal data via a third-party endpoint (https://mcp.neron.guru/mcp). Before installing or connecting:
- Verify the service and operator: ask for a homepage, privacy policy, and source code or repository. The registry metadata lists no homepage and the publisher is unknown — that reduces trust.
- Treat the Telegram token/password as high-value secrets: only request tokens you can revoke, and avoid pasting them into shared or cloud-backed config files. Prefer storing them in a secure credential store if possible.
- If possible, use a least-privilege token (read-only) for agents that only need to view data; avoid giving write/delete rights unless necessary.
- Understand that the skill allows raw Cypher queries and 'full' verbosity, which can return complete data dumps — don't grant it access to real sensitive data until you trust the service.
- Check TLS certificate and domain reputation for mcp.neron.guru and validate the Telegram bot identity (@NeronBetaBot) before sending credentials to it.
- Consider testing with throwaway or synthetic data first, and ensure you can revoke the token (/token) and that revocation invalidates prior tokens.
What would raise my confidence: a public homepage/privacy policy, audited source code or GitHub repo, documented token scopes, and a clear operator identity or third‑party audit.Like a lobster shell, security has layers — review code before you run it.
latestvk975wpvn7m0bpv0qe9qcjs0729832e15
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
