tengu_v3_intel_model_calibration
Live conformal-coverage telemetry: how often the model's stated 90% intervals actually contain the realised 5d returns. Built nightly over the trailing 30 days of prediction-outcome pairs. Returns stated_coverage (target, typically 0.90), realised_coverage (actual, e.g. 0.78), coverage_delta (gap...
This record as markdown: /tools/tengu-mcp/tengu-v3-intel-model-calibration.md
What tengu_v3_intel_model_calibration does on Tengu
AI agents call tengu_v3_intel_model_calibration to retrieve information from Tengu without modifying anything. It is typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Why tengu_v3_intel_model_calibration is rated Low
Even though tengu_v3_intel_model_calibration only reads data, uncontrolled read access leaks sensitive information and racks up API costs: an agent caught in a retry loop can make thousands of calls a minute without anyone noticing.
Attacks that exploit this kind of access
The rule that runs tengu_v3_intel_model_calibration safely
PolicyLayer is an MCP gateway: it sits between your AI agents and Tengu, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For tengu_v3_intel_model_calibration, this is the rule to start with:
tengu_v3_intel_model_calibration 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 Tengu, apply this rule, and every tengu_v3_intel_model_calibration call is checked against it from then on.
Questions about tengu_v3_intel_model_calibration
Live conformal-coverage telemetry: how often the model's stated 90% intervals actually contain the realised 5d returns. Built nightly over the trailing 30 days of prediction-outcome pairs. Returns stated_coverage (target, typically 0.90), realised_coverage (actual, e.g. 0.78), coverage_delta (gap, negative = under-covering), status (red/amber/green), n_pairs (sample size, ~110K typical), mean_interval_width_pct, mean_predicted_return_pct, mean_realised_return_pct, and an interpretation string. Treat status=red as a verdict-grade caveat — chat should attach 'model intervals currently under-covering' to any ml_prediction citation when this returns red. 1h cache. It is categorised as a Read tool in the Tengu MCP Server, which means it retrieves data without modifying state.
Register the Tengu MCP server in PolicyLayer and add a rule for tengu_v3_intel_model_calibration: 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 Tengu. Nothing to install.
tengu_v3_intel_model_calibration 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 tengu_v3_intel_model_calibration 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 tengu_v3_intel_model_calibration. 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.
tengu_v3_intel_model_calibration is provided by the Tengu MCP server (Hlobo-dev/tengu-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
More on Tengu, and thousands of servers like it.
This server
Across the catalogue