sandbox_ingest
Index arbitrary text into the context sandbox (FTS5). Raw content stays in SQLite — only a compact reference enters context. Use for large outputs, API responses, file contents, or any data you want searchable without flooding the context window.
This record as markdown: /tools/io-github-homenshum-nodebench/sandbox-ingest.md
What sandbox_ingest does on Nodebench
AI agents use sandbox_ingest to create or update resources in Nodebench, usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Nodebench environment.
Why sandbox_ingest is rated Medium
This tool writes/indexes data into a SQLite database (FTS5 full-text search index). It creates persistent records from arbitrary input, which is a Write operation. It does not execute code or delete data. Severity is medium because it accepts arbitrary text and stores it persistently, which could be misused to store sensitive or malicious content, but the blast radius is limited to the sandbox context.
From the tool's definition 'Index arbitrary text into the context sandbox (FTS5)' and 'Raw content stays in SQLite'
Attacks that exploit this kind of access
The rule that runs sandbox_ingest 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 sandbox_ingest, this is the rule to start with:
sandbox_ingest stays usable, but capped: an agent stuck in a loop can't make hundreds of changes a minute. 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 sandbox_ingest call is checked against it from then on.
Questions about sandbox_ingest
Index arbitrary text into the context sandbox (FTS5). Raw content stays in SQLite — only a compact reference enters context. Use for large outputs, API responses, file contents, or any data you want searchable without flooding the context window. It is categorised as a Write tool in the Nodebench MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Nodebench MCP server in PolicyLayer and add a rule for sandbox_ingest: 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.
sandbox_ingest is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the sandbox_ingest 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 sandbox_ingest. 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.
sandbox_ingest 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