Code Mode in Action

See how agents compose complex multi-step operations in a single tool call

Code Mode replaces hundreds of individual MCP tools with just two: search and execute. The agent writes JavaScript that runs in a sandboxed V8 isolate — and that code can fan out dozens of internal API calls, run them in parallel, aggregate results, and return a summary. All in one tool invocation.

Why It Works

LLMs have seen millions of lines of real-world JavaScript but only contrived tool-calling examples. Writing Promise.all() with a .map() is natural for them — chaining individual tool calls is not.

The sandbox provides:

• cnap.request() — typed API client with auth injected server-side
• Full JavaScript — Promise.all, Array.map, Date.now(), error handling, any logic the agent needs
• Up to 50 API calls per execution — enough for complex multi-step workflows
• No network escape — requests only reach the CNAP API, never the open internet

The agent’s code filters, aggregates, and formats data inside the sandbox before returning — so only the summary hits the context window, not raw API responses.

Examples

Parallel Log Analysis

35,000 lines across 11 pods in 506ms — parallel fan-out with in-sandbox aggregation

Resource Audit

CPU/memory capacity report in 56ms — unit parsing and computation from a single API call

Security Audit

9 security checks across 7 parallel API calls — then auto-remediates with network policies

Incident Debugging

Full SRE triage — pod health, events, error logs, rollout history — adapting at each step

Database Operations

Run SQL queries, check Redis memory, chain diagnostics — all through exec

Cross-Cluster Comparison

Compare versions, images, and CRDs across your entire fleet in one call

CRD Discovery

Discover, inspect, and interact with custom resources the agent has never seen

What Agents Can Do

Prompt	What happens
”Count log lines across all pods”	Parallel log fetches, line counting, sorted summary — 506ms
”Audit resource requests and limits”	CPU/memory unit parsing, per-pod aggregation — 56ms
”Audit the cluster for security”	9 checks across pods, RBAC, secrets, network policies
”Create network policies for my installs”	Discovers ports, generates and applies policies per namespace
”Why is my app returning 500s?”	Adaptive triage: pod health → events → logs → rollout history
”What’s the largest table in Postgres?”	Runs psql inside the container via exec
”Are my clusters on the same version?”	Parallel fan-out across all clusters, drift detection
”What CRDs are installed?”	Discovers custom APIs, reads schemas, queries instances

Each row is one or two tool calls. The agent writes the code, the sandbox executes it, and only the results reach the context window.

Related

Platform MCP

How Code Mode works and the two tools it exposes

Kubernetes Access

Kube API proxy, pod logs, and command execution