New Your team’s decisions, in one playbook every coding agent works from. Never answer your agent twice

project_search

Full-text (or regex) search across the project's source files (ripgrep-style). Read-only. Returns one match per line with its res:// path, 1-based line and column, and the matching text. Binary and oversized files are skipped; caches (.git/.godot/.import/node_modules) are never searched.

SERVERBreakpoint SOURCEbreakpoint-mcp
Low RISK CLASS
Category Read
Parameters 61 required
Recommended Allowedsee the rule below
Registry record Grade F, identity unverified Pull the record →

This record as markdown: /tools/breakpoint-mcp/project-search.md

What project_search does on Breakpoint

AI agents call project_search to retrieve information from Breakpoint without modifying anything. It is typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.

ParameterTypeRequiredDescription
path string Limit the search to a res:// or project-relative subdirectory (default: whole project)
query string Yes Text to find (literal by default; a regular expression when regex=true)
regex boolean Treat query as a JavaScript regular expression (default false = literal)
extensions array File extensions to include, without the dot (default: gd,cs,tscn,tres,gdshader,godot,cfg,json,md,txt,…)
ignore_case boolean Case-insensitive match (default false)
max_results integer Cap on returned matches (default 200)

Parameters from the server's own tool schema.

Why project_search is rated Low

This tool retrieves and queries data from source files without creating, modifying, deleting, or executing anything. The explicit "Read-only" designation and search-only functionality (ripgrep-style) confirm it as a Read category tool with low severity, as it has minimal blast radius - an agent using this tool can only discover information that already exists in the codebase.

From the tool's definition Tool description explicitly states "Read-only" and performs "Full-text (or regex) search across the project's source files" with no modification or side effects. Returns search results with file paths, line/column positions, and matching text.

Risk signalsAccepts file system path (path)

Questions about project_search

What does the project_search tool do? +

Full-text (or regex) search across the project's source files (ripgrep-style). Read-only. Returns one match per line with its res:// path, 1-based line and column, and the matching text. Binary and oversized files are skipped; caches (.git/.godot/.import/node_modules) are never searched. It is categorised as a Read tool in the Breakpoint MCP Server, which means it retrieves data without modifying state.

What parameters does project_search accept? +

project_search accepts 6 parameters: path, query, regex, extensions, ignore_case, max_results. Required: query. The full parameter table on this page comes from the server's own tool schema.

How do I enforce a policy on project_search? +

Register the Breakpoint MCP server in PolicyLayer and add a rule for project_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 Breakpoint. Nothing to install.

What risk level is project_search? +

project_search is a Read tool with low risk. Read-only tools are generally safe to allow by default.

Can I rate-limit project_search? +

Yes. Add a rate_limit block to the project_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.

How do I block project_search completely? +

Set action: deny in the PolicyLayer policy for project_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.

What MCP server provides project_search? +

project_search is provided by the Breakpoint MCP server (breakpoint-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.

More on Breakpoint, and thousands of servers like it.

// THE MCP REGISTRY

PolicyLayer tracks 44,603 MCP servers and 515,000+ tools.

Every server has a live record: who publishes it, whether it answers without auth, its risk grade, every tool classified, the recommended policy. This page is one line of Breakpoint's. Pull the full record:

Teams ship this data inside their own products. See what a licence covers →

// GET IN TOUCH

Have a question or want to learn more? Send us a message.

Message sent.

We'll get back to you soon.