AI agents call synaptex_search to retrieve information from Synaptex without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves or queries indexed documentation across multiple repositories. It performs semantic search, which is a read-only operation that returns matching results without modifying, creating, or deleting any data. No side effects or external operations are triggered. This aligns with the Read category for tools that retrieve or query data.
From the tool's definition Tool description: 'Recherche sémantique' (semantic search) in indexed CLAUDE.md files. The name contains 'search' and the description explicitly indicates querying/searching indexed data with no modification capability.
Attacks that exploit this kind of access
Recherche sémantique dans les CLAUDE.md indexés de tous les projets Forgejo. It is categorised as a Read tool in the Synaptex MCP Server, which means it retrieves data without modifying state.
Register the Synaptex MCP server in PolicyLayer and add a rule for synaptex_search: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Synaptex. Nothing to install.
synaptex_search is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the synaptex_search rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for synaptex_search. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
synaptex_search is provided by the Synaptex MCP server (jeromelabophy-gif/synaptex_claude). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →