convex_scan_capabilities
Analyze Convex source files for structural patterns. Scans a single file or entire convex/ directory and returns a capability report: function types (queries, mutations, actions, crons, http), schema constructs (tables, indexes, validators), data access (query, get, insert, patch, delete, collect...
This record as markdown: /tools/io-github-homenshum-nodebench/convex-scan-capabilities.md
What convex_scan_capabilities does on Nodebench
AI agents call convex_scan_capabilities to retrieve information from Nodebench without modifying anything. It is typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Why convex_scan_capabilities is rated Low
This is a pure analysis and introspection tool that examines source code to identify patterns and capabilities. The description explicitly states 'Pure regex — no LLM needed' and describes only scanning and reporting activities. It does not create, modify, delete, execute arbitrary code, or trigger external operations.
From the tool's definition "Analyze Convex source files for structural patterns. Scans a single file or entire convex/ directory and returns a capability report" — the tool performs static code analysis via regex patterns to extract and report information about function types, schema…
Attacks that exploit this kind of access
The rule that runs convex_scan_capabilities safely
PolicyLayer is an MCP gateway: it sits between your AI agents and Nodebench, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For convex_scan_capabilities, this is the rule to start with:
convex_scan_capabilities is read-only, so it stays allowed. Everything else on the server is denied unless you say otherwise.
The button opens the PolicyLayer dashboard: create your workspace, connect Nodebench, apply this rule, and every convex_scan_capabilities call is checked against it from then on.
Questions about convex_scan_capabilities
Analyze Convex source files for structural patterns. Scans a single file or entire convex/ directory and returns a capability report: function types (queries, mutations, actions, crons, http), schema constructs (tables, indexes, validators), data access (query, get, insert, patch, delete, collect, paginate), auth/context (getUserIdentity, runMutation, scheduler), storage (store, getUrl, delete), and client-side hooks (useQuery, useMutation, useAction). Pure regex — no LLM needed. It is categorised as a Read tool in the Nodebench MCP Server, which means it retrieves data without modifying state.
Register the Nodebench MCP server in PolicyLayer and add a rule for convex_scan_capabilities: 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 Nodebench. Nothing to install.
convex_scan_capabilities 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 convex_scan_capabilities 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 convex_scan_capabilities. 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.
convex_scan_capabilities is provided by the Nodebench MCP server (nodebench-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
More on Nodebench, and thousands of servers like it.
This server
Across the catalogue