Find code structurally similar to a snippet using AST fingerprinting. Parses the snippet and candidate code blocks into ASTs, extracts structural fingerprints, and computes Jaccard similarity. Args: project: Project name snippet: Code snippet to find similar code for language: Language of the sni...
AI agents call find_similar_code to retrieve information from Tree Sitter without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool only retrieves and analyzes code to identify structurally similar patterns. It performs read-only code analysis operations with no side effects, data modification, code execution, or resource consumption beyond computation. The blast radius of misuse is minimal—at worst, an AI agent could retrieve sensitive code patterns, but cannot modify, delete, or execute anything.
From the tool's definition Tool description states it 'Find[s] code structurally similar to a snippet' and 'Returns: List of similar code blocks with similarity scores'.
Documented attack patterns abuse exactly the kind of access find_similar_code gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Tree Sitter, and nothing reaches the server without passing your rules. This is the rule we recommend for find_similar_code:
{
"version": "1",
"default": "deny",
"tools": {
"find_similar_code": {}
}
}find_similar_code is read-only, so it stays allowed — but everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Find code structurally similar to a snippet using AST fingerprinting. Parses the snippet and candidate code blocks into ASTs, extracts structural fingerprints, and computes Jaccard similarity. Args: project: Project name snippet: Code snippet to find similar code for language: Language of the snippet (required) threshold: Minimum Jaccard similarity (0.0-1.0, default 0.6) max_results: Maximum number of results Returns: List of similar code blocks with similarity scores. It is categorised as a Read tool in the Tree Sitter MCP Server, which means it retrieves data without modifying state.
Register the Tree Sitter MCP server in PolicyLayer and add a rule for find_similar_code: 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 Tree Sitter. Nothing to install.
find_similar_code 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 find_similar_code 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 find_similar_code. 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.
find_similar_code is provided by the Tree Sitter MCP server (wrale/mcp-server-tree-sitter). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 26 Tree Sitter tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
26 Tree Sitter tools catalogued and risk-classified — across an index of 42,500+ MCP servers.