Hodge-style decomposition of validation failures. Splits the error between original and corrected values into three orthogonal components: - Exact: direct rule violations (a field breaks a specific math rule) - Co-exact: constraint boundary violations (a field is at the edge of valid range) - Har...
Risk signalsHandles credentials or secrets (api_key)
Part of the Governance Platform server.
Free to start. No card required.
AI agents call decompose_failure to retrieve information from Governance Platform without modifying any data. This is common in research, monitoring, and reporting workflows where the agent needs context before taking action. Because read operations don't change state, they are generally safe to allow without restrictions -- but you may still want rate limits to control API costs.
Even though decompose_failure only reads data, uncontrolled read access can leak sensitive information or rack up API costs. An agent caught in a retry loop could make thousands of calls per minute. A rate limit gives you a safety net without blocking legitimate use.
Read-only tools are safe to allow by default. No rate limit needed unless you want to control costs.
{
"version": "1",
"default": "deny",
"tools": {
"decompose_failure": {}
}
}See the full Governance Platform policy for all 31 tools.
These attack patterns abuse exactly the kind of access decompose_failure gives an agent. Each links to the full case and the policy that stops it:
Other read tools across the catalogue. The same approach applies to each: allow, with a rate cap to control cost.
Hodge-style decomposition of validation failures. Splits the error between original and corrected values into three orthogonal components: - Exact: direct rule violations (a field breaks a specific math rule) - Co-exact: constraint boundary violations (a field is at the edge of valid range) - Harmonic: systemic structural errors (the overall data shape is wrong) Different from analyze_anomaly: analyze_anomaly explains why a single payload is anomalous given the platform's structural fingerprints. decompose_failure compares two known states (an original and its correction) and tells you which kind of failure accounts for the difference. Use decompose_failure when you have a known-correct version to diff against; use analyze_anomaly when you only have the suspicious payload. Use this after a validation failure to understand the structural nature of the error, not just which fields differ. Args: api_key: GeodesicAI API key (starts with gai_) original_values: The original field values (numeric key-value pairs) corrected_values: The corrected/expected field values derivation_rules: Math rules. Optional if blueprint provided. formal_constraints: Constraints. Optional if blueprint provided. blueprint: Load rules from this Blueprint instead of passing them inline. Caller must own the Blueprint. Returns: primary_cause: one of "exact" / "co_exact" / "harmonic" exact: details of direct rule violations co_exact: details of constraint boundary violations harmonic: details of systemic structural errors contributing_fields: map of field_name → contribution magnitude total_error_magnitude: scalar L2 norm of the full error. It is categorised as a Read tool in the Governance Platform MCP Server, which means it retrieves data without modifying state.
Register the Governance Platform MCP server in PolicyLayer and add a rule for decompose_failure: 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 Governance Platform. Nothing to install.
decompose_failure 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 decompose_failure 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 decompose_failure. 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.
decompose_failure is provided by the Governance Platform MCP server (https://app.geodesiclabs.ai/mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 31 Governance Platform tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
4,600+ MCP servers and 31,000+ tools scanned and risk-classified.