Performs full-text search on HexDocs documentation using Typesense search engine. This searches the actual documentation content across all packages on HexDocs. Note: If TideWave is available in your current Phoenix project, prefer using TideWave's documentation search for better project context....
Risk signalsAccepts freeform code/query input (query)
Part of the HexDocs MCP server.
Free to start. No card required.
AI agents call fulltext_search to retrieve information from HexDocs MCP without modifying any data. This is common in research, monitoring, and reporting workflows where the agent needs context before taking action. Because read operations don't change state, they are generally safe to allow without restrictions -- but you may still want rate limits to control API costs.
Even though fulltext_search only reads data, uncontrolled read access can leak sensitive information or rack up API costs. An agent caught in a retry loop could make thousands of calls per minute. A rate limit gives you a safety net without blocking legitimate use.
Read-only tools are safe to allow by default. No rate limit needed unless you want to control costs.
{
"version": "1",
"default": "deny",
"tools": {
"fulltext_search": {}
}
} See the full HexDocs MCP policy for all 5 tools.
These attack patterns abuse exactly the kind of access fulltext_search gives an agent. Each links to the full case and the policy that stops it:
Other read tools across the catalogue. The same approach applies to each: allow, with a rate cap to control cost.
Performs full-text search on HexDocs documentation using Typesense search engine. This searches the actual documentation content across all packages on HexDocs. Note: If TideWave is available in your current Phoenix project, prefer using TideWave's documentation search for better project context. Use this tool when TideWave is not available or for searching packages not in your current project. Query Syntax: - Basic search: "Phoenix.LiveView" (searches for both terms) - Exact phrase: "\"handle event\"" (use escaped quotes) - AND operator: "Phoenix AND LiveView" (both terms required) - OR operator: "Phoenix OR Plug" (either term) - Exclude terms: "Phoenix -test" (minus sign excludes) - Module/function: "Enum.map" or "GenServer.handle_call" Best Practices: - Use exact module.function names for precise results - Combine with packageName filter for focused search - Use quotes for multi-word exact phrases - For callbacks use patterns like "@callback handle_" - For types use patterns like "@type t()" Examples: - Find LiveView event handlers: query: "handle_event", packageName: "phoenix_live_view" - Find Ecto changesets: query: "changeset", packageName: "ecto" - Find specific function: query: "\"Enum.map/2\"" - Find type definitions: query: "@type", packageName: "phoenix". It is categorised as a Read tool in the HexDocs MCP MCP Server, which means it retrieves data without modifying state.
Register the HexDocs MCP server in PolicyLayer and add a rule for fulltext_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 HexDocs MCP. Nothing to install.
fulltext_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 fulltext_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 fulltext_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.
fulltext_search is provided by the HexDocs MCP server (bradleygolden/hexdocs-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 5 HexDocs MCP tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
4,600+ MCP servers and 31,000+ tools scanned and risk-classified.