Route a Call to the point on the price-performance frontier that meets your Quality Gate, up or down. A Call is one inference request through Seracade. Seracade also accepts native Anthropic (/v1/messages), Google (/v1beta/...), and xAI request shapes directly when the customer's SDK targets them...
Handles credentials or secrets (api_key)
Part of the Seracade MCP server. Enforce policies on this tool with Intercept, the open-source MCP proxy.
AI agents invoke seracade_route to trigger processes or run actions in Seracade. Execute operations can have side effects beyond the immediate call -- triggering builds, sending notifications, or starting workflows. Rate limits and argument validation are essential to prevent runaway execution.
seracade_route can trigger processes with real-world consequences. An uncontrolled agent might start dozens of builds, send mass notifications, or kick off expensive compute jobs. Intercept enforces rate limits and validates arguments to keep execution within safe bounds.
Execute tools trigger processes. Rate-limit and validate arguments to prevent unintended side effects.
tools:
seracade_route:
rules:
- action: allow
rate_limit:
max: 10
window: 60
validate:
required_args: trueSee the full Seracade policy for all 3 tools.
Agents calling execute-class tools like seracade_route have been implicated in these attack patterns. Read the full case and prevention policy for each:
Other tools in the Execute risk category across the catalogue. The same policy patterns (rate-limit, validate) apply to each.
seracade_route is one of the high-risk operations in Seracade. For the full severity-focused view — only the high-risk tools with their recommended policies — see the breakdown for this server, or browse all high-risk tools across every MCP server.
Route a Call to the point on the price-performance frontier that meets your Quality Gate, up or down. A Call is one inference request through Seracade. Seracade also accepts native Anthropic (/v1/messages), Google (/v1beta/...), and xAI request shapes directly when the customer's SDK targets them; cross-shape substitution translates the response back to the native shape so the customer SDK keeps working unchanged (non-streaming and streaming text plus tool calls; multimodal inputs and provider-specific features pass through). Cross-shape routing is opt-in per customer (POST /api/cross-shape-routing { enabled: true }). After signup the customer activates routing on the dashboard with one Start Routing click that applies across the task type rows they confirm; per-row pause is one click on any row and takes effect within one second. Until the click, the call passes through to the customer's default unchanged. Decisions on activated task types the customer has audited are grounded in empirical Quality Scores measured on the customer's own traffic; decisions on activated task types not yet audited route from Seracade's internal benchmarks at the customer's per-task quality threshold. Each Routing Decision returned in the response carries a source field ("v0_public", "audit", or "reauditor") and a confidence_tier ("high" or "low") so the caller can see which layer wrote the decision and how strong the evidence is. Activate all confirmed rows via POST /api/routing/activate-all (or per row via POST /api/routing/applied with task_type and enabled=true). Pause any row with /api/routing/applied and enabled=false. Returns the full completion, the model used, the original model, the per-Call price delta, and the source. Optional Step headers (X-Seracade-Step-Id, X-Seracade-Parent-Call-Id, X-Seracade-Trace-Id) identify position in an agent trajectory and are persisted on the Call record. Recommended: set SERACADE_API_KEY env var instead of passing api_key inline.. It is categorised as a Execute tool in the Seracade MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Add a rule in your Intercept YAML policy under the tools section for seracade_route. You can allow, deny, rate-limit, or validate arguments. Then run Intercept as a proxy in front of the Seracade MCP server.
seracade_route is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the seracade_route rule in your Intercept 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 Intercept policy for seracade_route. 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.
seracade_route is provided by the Seracade MCP server (seracade). Intercept sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic policy on every MCP tool call. Per-identity grants. Full audit log.