Agent KnowledgeSAFE
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
Complete knowledge management for AI assistants MCP server with Elasticsearch search and document management.
[](https://python.org) [](https://modelcontextprotocol.io) [](LICENSE)
🚀 Features
🔑 All-in-One Solution:
- 🔍 Elasticsearch: Search, index, and manage documents
- 📊 Document Validation: Schema-enforced structure
- ⚙️ Configuration: Complete config management
- 🛡️ Security: Sandboxed operations
✨ Benefits:
- 🎯 20 Tools for knowledge management
- 🤖 Works with any MCP-compatible AI (Claude, ChatGPT, VS Code, etc.)
- 📚 Smart document management with validation
- ⚡ Elasticsearch integration for powerful search
⚡ Quick Start
Installation
# Install with uvx (recommended) uvx agent-knowledge-mcp
Setup for Claude Desktop
Add to claude_desktop_config.json:
{
"mcpServers": {
"agent-knowledge": {
"command": "uvx",
"args": ["agent-knowledge-mcp"]
}
}
}Setup for VS Code
[](https://insiders.vscode.dev/redirect/mcp/install?name=agent-knowledge&inputs=%5B%5D&config=%7B%22command%22%3A%22uvx%22%2C%22args%22%3A%5B%22agent-knowledge-mcp%22%5D%7D)
🛠️ What You Can Do
Try these with your AI assistant:
- "Search documents for API authentication info"
- "Index this document with proper tags"
- "Create API documentation template"
- "Find related documents on specific topics"
- *"Up
318c3e651607OBSERVED · 2026-10-08Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control.
claude mcp add agent-knowledge-mcp -- uvx agent-knowledge-mcp
{
"mcpServers": {
"agent-knowledge-mcp": {
"command": "uvx",
"args": [
"agent-knowledge-mcp"
]
}
}
}Exposed tools (10)
8 read · 0 write · 2 destructive. Blast radius: 2 tools can delete or overwrite — an agent that can be talked into calling a tool can be talked into calling this one.
| Tool | Risk | Description |
|---|---|---|
delete_index | destructive | Delete an Elasticsearch index permanently. |
delete_index_metadata | destructive | Delete metadata documentation for an Elasticsearch index. |
elasticsearch_status | read | Check comprehensive status of Elasticsearch and Kibana Docker containers. |
get_config | read | Get the complete configuration from config.json file. |
list_indices | read | List all available Elasticsearch indices with basic statistics. |
reload_config | read | Reload configuration and reinitialize all system components with updated settings. |
server_status | read | Check comprehensive server status including version, installation method, and update availability. |
server_upgrade | read | Upgrade the MCP server when installed via uvx with comprehensive backup and restoration. |
validate_config | read | Validate configuration with comprehensive structure checking and detailed feedback. |
validate_document_schema | read | Validate document structure against knowledge base schema standards. |
Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | PASS |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | WARN |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (1 observation(s))
- Network
- declared (4 observation(s))
- Shell
- none-observed
- Dependencies
- not all pinned
- Secrets in source
- none-found
Findings (5)
delete_index, delete_index_metadata
content_hash = hashlib.md5(content.encode()).hexdigest()[:8]
timestamp_hash = hashlib.md5(str(time.time()).encode()).hexdigest()[:8]
mcp, fastmcp, pydantic, elasticsearch, docker, requests
🚨 MANDATORY: Search knowledge base with `search` tool for relevant information - This is REQUIRED for EVERY user response - NO EXCEPTIONS - failure to search = workflow violation
Gates applied: no_behavioural_pass.
318c3e651607full audit observations/trust-audit/mcp-server/itshare4u__agent-knowledge.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-08 | 318c3e651607 | SAFE | B | 89 | first audit |
Questions
What tools does Agent Knowledge expose?
10 in total: 8 read-only, 0 that write, and 2 that can delete or overwrite (delete_index, delete_index_metadata). Every one is listed on this page with its risk.
Is Agent Knowledge safe to connect to an agent?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean server reads B. Separately from the audit: 2 of its tools can destroy data, so scope the token you give it to what you actually need.
What credentials does Agent Knowledge need?
No credential environment variables were found in its source, so it appears to need none.
How current is this page?
The grade is for one exact copy of the source (318c3e651607), read on 2026-10-08. The repository is watched and re-audited when it changes.