governance-platform

Validate structured data against a Blueprint's rules BEFORE the result is used. Returns PASS, FAIL, or REVIEW with plain-language findings, repair suggestions, a determinism hash, and a re-verifiable certificate. Same input + same rules = same verdict, every time.

{
  "type": "object",
  "properties": {
    "api_key": {
      "description": "GeodesicAI API key (gai_...)",
      "title": "Api Key",
      "type": "string"
    },
    "structured_data": {
      "additionalProperties": true,
      "description": "The document's extracted fields as key/value pairs. Keys are open by design - your Blueprint's rules define what is checked",
      "title": "Structured Data",
      "type": "object"
    },
    "blueprint": {
      "default": "default",
      "description": "Blueprint name (workflow_name) to use",
      "title": "Blueprint",
      "type": "string"
    }
  },
  "required": [
    "api_key",
    "structured_data"
  ],
  "title": "validateArguments"
}

Validate structured data against a Blueprint and, when it fails, include repair suggestions (corrected values with the rule each fix is based on) in the same call. Same verdicts as validate: PASS, FAIL, or REVIEW, with reasons and proof.

{
  "type": "object",
  "properties": {
    "api_key": {
      "description": "GeodesicAI API key (gai_...)",
      "title": "Api Key",
      "type": "string"
    },
    "structured_data": {
      "additionalProperties": true,
      "description": "The document's extracted fields as key/value pairs. Keys are open by design - your Blueprint's rules define what is checked",
      "title": "Structured Data",
      "type": "object"
    },
    "blueprint": {
      "default": "default",
      "description": "Blueprint name (workflow_name) to use",
      "title": "Blueprint",
      "type": "string"
    }
  },
  "required": [
    "api_key",
    "structured_data"
  ],
  "title": "validate_repairArguments"
}

Create a Blueprint - the governance contract validation runs against. A Blueprint defines what correct means for your data: fields, the math that must hold between them, and acceptable ranges. Start from load_rule_pack or discover_patterns if you have no rules yet; invoke the blueprint_guide prompt for the full rule/constraint reference. Returns the new Blueprint's API key.

{
  "type": "object",
  "properties": {
    "api_key": {
      "description": "GeodesicAI API key (gai_...)",
      "title": "Api Key",
      "type": "string"
    },
    "customer_name": {
      "description": "Organization or project name (also used for storage folder naming)",
      "title": "Customer Name",
      "type": "string"
    },
    "workflow_name": {
      "description": "Unique Blueprint identifier; the value passed as 'blueprint' in validate",
      "title": "Workflow Name",
      "type": "string"
    },
    "mode": {
      "default": "observe",
      "description": "observe: platform checks the agent's work; enforce: platform computes derived fields itself",
      "enum": [
        "observe",
        "enforce"
      ],
      "title": "Mode",
      "type": "string"
    },
    "extracted_fields": {
      "default": null,
      "description": "Field names the agent extracts from source data, e.g. ['vendor','qty','unit_cost']",
      "items": {},
      "title": "Extracted Fields",
      "type": "array"
    },
    "derived_fields": {
      "default": null,
      "description": "Field names the platform computes from other fields, e.g. ['subtotal','total']",
      "items": {},
      "title": "Derived Fields",
      "type": "array"
    },
    "derivation_rules": {
      "default": null,
      "description": "Math rules as objects. Types: add, subtract, multiply, divide, round, copy, sum (multi-operand), items_multiply, items_sum. Each needs 'type' plus its fields; see the blueprint_guide prompt",
      "items": {},
      "title": "Derivation Rules",
      "type": "array"
    },
    "formal_constraints": {
      "default": null,
      "description": "Constraint objects. Types incl. magnitude_anchor {field,min,max}, relative_anchor {field,reference_field,ratio_min,ratio_max}, max_action_threshold {field,threshold,on_violation}, required_fields {fields}, equals, range, in_set, regex_match, items_magnitude_anchor; see the blueprint_guide prompt",
      "items": {},
      "title": "Formal Constraints",
      "type": "array"
    },
    "semantic_checks": {
      "default": null,
      "description": "Domain-specific semantic check objects",
      "items": {},
      "title": "Semantic Checks",
      "type": "array"
    },
    "require_math": {
      "default": true,
      "description": "Validate mathematical relationships",
      "title": "Require Math",
      "type": "boolean"
    },
    "require_consistency": {
      "default": true,
      "description": "Check internal field consistency",
      "title": "Require Consistency",
      "type": "boolean"
    },
    "require_coherence": {
      "default": true,
      "description": "Check cross-field plausibility",
      "title": "Require Coherence",
      "type": "boolean"
    },
    "require_provenance": {
      "default": false,
      "description": "Require extraction source locations for fields",
      "title": "Require Provenance",
      "type": "boolean"
    },
    "require_high_assurance": {
      "default": false,
      "description": "Strictest mode: every check must pass",
      "title": "Require High Assurance",
      "type": "boolean"
    },
    "enable_anomaly_detection": {
      "default": false,
      "description": "Flag records that break no rules but do not fit the reference pattern",
      "title": "Enable Anomaly Detection",
      "type": "boolean"
    },
    "enable_drift_tracking": {
      "default": false,
      "description": "Track pattern stability across batches",
      "title": "Enable Drift Tracking",
      "type": "boolean"
    }
  },
  "required": [
    "api_key",
    "customer_name",
    "workflow_name"
  ],
  "title": "create_blueprintArguments"
}

List the Blueprints on this account with field/rule/constraint counts and mode. Use the returned workflow_name as 'blueprint' in validate.

{
  "type": "object",
  "properties": {
    "api_key": {
      "description": "GeodesicAI API key (gai_...)",
      "title": "Api Key",
      "type": "string"
    }
  },
  "required": [
    "api_key"
  ],
  "title": "list_blueprintsArguments"
}

One-shot repair: return corrected values that would make failing data valid under the Blueprint. Use repair_path to see the steps instead.

{
  "type": "object",
  "properties": {
    "api_key": {
      "description": "GeodesicAI API key (gai_...)",
      "title": "Api Key",
      "type": "string"
    },
    "structured_data": {
      "additionalProperties": true,
      "description": "The document's extracted fields as key/value pairs. Keys are open by design - your Blueprint's rules define what is checked",
      "title": "Structured Data",
      "type": "object"
    },
    "derivation_rules": {
      "default": null,
      "description": "Math rules as objects. Types: add, subtract, multiply, divide, round, copy, sum (multi-operand), items_multiply, items_sum. Each needs 'type' plus its fields; see the blueprint_guide prompt",
      "items": {},
      "title": "Derivation Rules",
      "type": "array"
    },
    "formal_constraints": {
      "default": null,
      "description": "Constraint objects. Types incl. magnitude_anchor {field,min,max}, relative_anchor {field,reference_field,ratio_min,ratio_max}, max_action_threshold {field,threshold,on_violation}, required_fields {fields}, equals, range, in_set, regex_match, items_magnitude_anchor; see the blueprint_guide prompt",
      "items": {},
      "title": "Formal Constraints",
      "type": "array"
    },
    "blueprint": {
      "default": null,
      "description": "Blueprint name (workflow_name) to use",
      "title": "Blueprint",
      "type": "string"
    }
  },
  "required": [
    "api_key",
    "structured_data"
  ],
  "title": "repairArguments"
}

Static pre-deploy analysis of a Blueprint's rule set. Returns a health verdict - healthy, acceptable, fragile, rigid, split, brittle_islands, or unsatisfiable - with advice, including joint conflicts pairwise checks miss.

{
  "type": "object",
  "properties": {
    "api_key": {
      "description": "GeodesicAI API key (gai_...)",
      "title": "Api Key",
      "type": "string"
    },
    "blueprint": {
      "default": "",
      "description": "Blueprint name (workflow_name) to use",
      "title": "Blueprint",
      "type": "string"
    },
    "config": {
      "additionalProperties": true,
      "default": null,
      "description": "Raw blueprint config with derivation_rules and formal_constraints (used when 'blueprint' is not given)",
      "title": "Config",
      "type": "object"
    }
  },
  "required": [
    "api_key"
  ],
  "title": "check_blueprint_healthArguments"
}

Compare two payloads under the dual-hash design: content_hash is content_hash normalizes field order and numeric formatting. Semantic comparison preserves field roles; renaming requires an explicit bijection. With supplied rules, scalar types and whitespace remain significant. Structural similarity alone does not establish decision equivalence.

{
  "type": "object",
  "properties": {
    "api_key": {
      "description": "GeodesicAI API key (gai_...)",
      "title": "Api Key",
      "type": "string"
    },
    "payload_a": {
      "additionalProperties": true,
      "description": "First structured payload (arbitrary JSON object)",
      "title": "Payload A",
      "type": "object"
    },
    "payload_b": {
      "additionalProperties": true,
      "description": "Second structured payload to compare against payload_a",
      "title": "Payload B",
      "type": "object"
    },
    "field_mapping": {
      "additionalProperties": true,
      "default": null,
      "description": "Explicit one-to-one field renaming from A to B",
      "title": "Field Mapping",
      "type": "object"
    },
    "rules_a": {
      "default": null,
      "description": "Derivation rules for A",
      "items": {},
      "title": "Rules A",
      "type": "array"
    },
    "rules_b": {
      "default": null,
      "description": "Derivation rules for B",
      "items": {},
      "title": "Rules B",
      "type": "array"
    },
    "constraints_a": {
      "default": null,
      "description": "Formal constraints for A",
      "items": {},
      "title": "Constraints A",
      "type": "array"
    },
    "constraints_b": {
      "default": null,
      "description": "Formal constraints for B",
      "items": {},
      "title": "Constraints B",
      "type": "array"
    }
  },
  "required": [
    "api_key",
    "payload_a",
    "payload_b"
  ],
  "title": "compare_semantic_equivalenceArguments"
}

Quality-govern an in-progress AI generation step BEFORE its output is used (complements validate, which checks finished documents). Returns an action - STOP, CONTINUE, REPAIR_REGION, REUSE_MOTIF, REVIEW, ESCALATE - with a plain-language explanation. Structural scores do not establish task correctness. Check safe_to_finalize and acceptance coverage. Persistence success is reported; read traces in the same Blueprint namespace.

{
  "type": "object",
  "properties": {
    "api_key": {
      "description": "GeodesicAI API key (gai_...)",
      "title": "Api Key",
      "type": "string"
    },
    "task_type": {
      "description": "Kind of generation step being governed",
      "enum": [
        "generative_text",
        "retrieval",
        "generic"
      ],
      "title": "Task Type",
      "type": "string"
    },
    "payload": {
      "additionalProperties": true,
      "description": "Task-type payload: generative_text {text,...}; retrieval {query,candidates}; generic {features}",
      "title": "Payload",
      "type": "object"
    },
    "inference_id": {
      "description": "Caller-chosen ID grouping the steps of one generation",
      "title": "Inference Id",
      "type": "string"
    },
    "constraints": {
      "additionalProperties": true,
      "default": null,
      "description": "Optional governance constraint config object",
      "title": "Constraints",
      "type": "object"
    },
    "source": {
      "default": "mcp",
      "description": "Free-form caller label recorded for audit",
      "title": "Source",
      "type": "string"
    },
    "step_index": {
      "default": 0,
      "description": "Step number within this generation (0-based)",
      "title": "Step Index",
      "type": "integer"
    },
    "blueprint": {
      "default": "default",
      "description": "Owned Blueprint namespace for the trace",
      "title": "Blueprint",
      "type": "string"
    }
  },
  "required": [
    "api_key",
    "task_type",
    "payload",
    "inference_id"
  ],
  "title": "govern_inferenceArguments"
}

Retrieve the durable audit trail for a governed generation: every recorded decision and its reasons.

{
  "type": "object",
  "properties": {
    "api_key": {
      "description": "GeodesicAI API key (gai_...)",
      "title": "Api Key",
      "type": "string"
    },
    "inference_id": {
      "description": "Caller-chosen ID grouping the steps of one generation",
      "title": "Inference Id",
      "type": "string"
    },
    "blueprint": {
      "default": "default",
      "description": "Blueprint namespace used when recording the trace",
      "title": "Blueprint",
      "type": "string"
    },
    "blueprint_version": {
      "default": "",
      "description": "Optional historical blueprint_version hash returned by govern_inference",
      "title": "Blueprint Version",
      "type": "string"
    }
  },
  "required": [
    "api_key",
    "inference_id"
  ],
  "title": "get_inference_traceArguments"
}

Recent generation-governance decisions in this owner's Blueprint version.

{
  "type": "object",
  "properties": {
    "api_key": {
      "description": "GeodesicAI API key (gai_...)",
      "title": "Api Key",
      "type": "string"
    },
    "limit": {
      "default": 25,
      "description": "Maximum rows to return",
      "title": "Limit",
      "type": "integer"
    },
    "action": {
      "default": "",
      "description": "Optional action filter (STOP, CONTINUE, REVIEW, ...)",
      "title": "Action",
      "type": "string"
    },
    "blueprint": {
      "default": "default",
      "description": "Blueprint namespace used when recording the trace",
      "title": "Blueprint",
      "type": "string"
    },
    "blueprint_version": {
      "default": "",
      "description": "Optional historical blueprint_version hash returned by govern_inference",
      "title": "Blueprint Version",
      "type": "string"
    }
  },
  "required": [
    "api_key"
  ],
  "title": "recent_inference_decisionsArguments"
}

Independently re-verify a validation certificate. Integrity mode checks the hash chain; full mode (certificate + original data) recomputes every attested rule from scratch - trust nothing, recheck everything.

{
  "type": "object",
  "properties": {
    "api_key": {
      "description": "GeodesicAI API key (gai_...)",
      "title": "Api Key",
      "type": "string"
    },
    "certificate": {
      "additionalProperties": true,
      "description": "The certificate object from a prior validation response",
      "title": "Certificate",
      "type": "object"
    },
    "data": {
      "additionalProperties": true,
      "default": null,
      "description": "Original payload for full re-verification; omit for integrity-only mode",
      "title": "Data",
      "type": "object"
    },
    "derivation_rules": {
      "default": null,
      "description": "Rule list for independent attestation recomputation (optional)",
      "items": {},
      "title": "Derivation Rules",
      "type": "array"
    },
    "formal_constraints": {
      "default": null,
      "description": "Optional constraints to match against the committed bundle",
      "items": {},
      "title": "Formal Constraints",
      "type": "array"
    }
  },
  "required": [
    "api_key",
    "certificate"
  ],
  "title": "verify_certificateArguments"
}

Sweep the Blueprint's numeric constraint bounds and report verdict stability: the stable band, the scales where the verdict first flips, and advice. Use before deploying bound changes.

{
  "type": "object",
  "properties": {
    "api_key": {
      "description": "GeodesicAI API key (gai_...)",
      "title": "Api Key",
      "type": "string"
    },
    "blueprint": {
      "default": "",
      "description": "Blueprint name (workflow_name) to use",
      "title": "Blueprint",
      "type": "string"
    },
    "config": {
      "additionalProperties": true,
      "default": null,
      "description": "Raw blueprint config to profile (used when 'blueprint' is not given)",
      "title": "Config",
      "type": "object"
    }
  },
  "required": [
    "api_key"
  ],
  "title": "profile_blueprint_robustnessArguments"
}

Deterministic forward reasoning: from the current data state, generate and rank the valid next states reachable under the Blueprint's rules.

{
  "type": "object",
  "properties": {
    "api_key": {
      "description": "GeodesicAI API key (gai_...)",
      "title": "Api Key",
      "type": "string"
    },
    "structured_data": {
      "additionalProperties": true,
      "description": "The document's extracted fields as key/value pairs. Keys are open by design - your Blueprint's rules define what is checked",
      "title": "Structured Data",
      "type": "object"
    },
    "blueprint": {
      "default": "default",
      "description": "Blueprint name (workflow_name) to use",
      "title": "Blueprint",
      "type": "string"
    },
    "max_depth": {
      "default": 3,
      "description": "Search depth, 1-10",
      "title": "Max Depth",
      "type": "integer"
    },
    "max_branches": {
      "default": 5,
      "description": "Branches per step, 1-10",
      "title": "Max Branches",
      "type": "integer"
    },
    "rank_by": {
      "default": "drift",
      "description": "Ranking criterion for returned paths",
      "enum": [
        "drift",
        "confidence",
        "shortest",
        "risk"
      ],
      "title": "Rank By",
      "type": "string"
    }
  },
  "required": [
    "api_key",
    "structured_data"
  ],
  "title": "forecastArguments"
}

Learn candidate validation rules and structural document types from a batch of your records, deterministically - no Blueprint required. Promote results with approve_rule. Source data is not stored.

{
  "type": "object",
  "properties": {
    "api_key": {
      "description": "GeodesicAI API key (gai_...)",
      "title": "Api Key",
      "type": "string"
    },
    "documents": {
      "description": "List of structured records (objects) to analyze",
      "items": {},
      "title": "Documents",
      "type": "array"
    },
    "blueprint": {
      "default": "default",
      "description": "Blueprint name (workflow_name) to use",
      "title": "Blueprint",
      "type": "string"
    }
  },
  "required": [
    "api_key",
    "documents"
  ],
  "title": "discover_patternsArguments"
}

Find the shortest sequence of field changes taking invalid data to a valid state, as an ordered path of intermediate states. Different from repair (one-shot nearest fix): use repair_path to explain or audit the fix, or compare alternative repairs.

{
  "type": "object",
  "properties": {
    "api_key": {
      "description": "GeodesicAI API key (gai_...)",
      "title": "Api Key",
      "type": "string"
    },
    "structured_data": {
      "additionalProperties": true,
      "description": "The document's extracted fields as key/value pairs. Keys are open by design - your Blueprint's rules define what is checked",
      "title": "Structured Data",
      "type": "object"
    },
    "blueprint": {
      "default": "default",
      "description": "Blueprint name (workflow_name) to use",
      "title": "Blueprint",
      "type": "string"
    },
    "max_depth": {
      "default": 4,
      "description": "Search depth, 1-10",
      "title": "Max Depth",
      "type": "integer"
    },
    "rank_by": {
      "default": "shortest",
      "description": "Ranking criterion for returned paths",
      "enum": [
        "shortest",
        "drift",
        "confidence",
        "risk"
      ],
      "title": "Rank By",
      "type": "string"
    }
  },
  "required": [
    "api_key",
    "structured_data"
  ],
  "title": "repair_pathArguments"
}

Run the same data under two rule sets and compare which future states remain valid - what-if analysis for rule changes.

{
  "type": "object",
  "properties": {
    "api_key": {
      "description": "GeodesicAI API key (gai_...)",
      "title": "Api Key",
      "type": "string"
    },
    "structured_data": {
      "additionalProperties": true,
      "description": "The document's extracted fields as key/value pairs. Keys are open by design - your Blueprint's rules define what is checked",
      "title": "Structured Data",
      "type": "object"
    },
    "blueprint": {
      "default": "default",
      "description": "Blueprint name (workflow_name) to use",
      "title": "Blueprint",
      "type": "string"
    },
    "rules_b": {
      "default": null,
      "description": "Alternative derivation rules (rule set B)",
      "items": {},
      "title": "Rules B",
      "type": "array"
    },
    "constraints_b": {
      "default": null,
      "description": "Alternative constraints (rule set B)",
      "items": {},
      "title": "Constraints B",
      "type": "array"
    }
  },
  "required": [
    "api_key",
    "structured_data"
  ],
  "title": "counterfactualArguments"
}

Explain whether a record fits the usual pattern for records like it, and which fields stand out. No Blueprint required.

{
  "type": "object",
  "properties": {
    "api_key": {
      "description": "GeodesicAI API key (gai_...)",
      "title": "Api Key",
      "type": "string"
    },
    "structured_data": {
      "additionalProperties": true,
      "description": "The document's extracted fields as key/value pairs. Keys are open by design - your Blueprint's rules define what is checked",
      "title": "Structured Data",
      "type": "object"
    },
    "blueprint": {
      "default": "default",
      "description": "Discovery namespace used by discover_patterns",
      "title": "Blueprint",
      "type": "string"
    }
  },
  "required": [
    "api_key",
    "structured_data"
  ],
  "title": "analyze_anomalyArguments"
}

Create a multi-agent sequential chain: stages validate in order against one Blueprint, repairs propagate forward, TTL bounds the run. Siblings: submit_chain_stage advances the chain; handoff_audit verifies a transition between stages. Returns chain_id.

{
  "type": "object",
  "properties": {
    "api_key": {
      "description": "GeodesicAI API key (gai_...)",
      "title": "Api Key",
      "type": "string"
    },
    "blueprint": {
      "description": "Blueprint governing all stages of the chain",
      "title": "Blueprint",
      "type": "string"
    },
    "stages": {
      "description": "Stage definitions, e.g. [{'stage_name':'extract','agent_name':'PDF Agent'}]; minimum 2",
      "items": {},
      "title": "Stages",
      "type": "array"
    },
    "ttl": {
      "default": 3600,
      "description": "Chain timeout in seconds; stages cannot advance after expiry",
      "title": "Ttl",
      "type": "integer"
    }
  },
  "required": [
    "api_key",
    "blueprint",
    "stages"
  ],
  "title": "create_chainArguments"
}

Submit data for the chain's current stage; the platform validates it and advances the chain if it passes. Response includes next-stage info and accumulated repairs.

{
  "type": "object",
  "properties": {
    "api_key": {
      "description": "GeodesicAI API key (gai_...)",
      "title": "Api Key",
      "type": "string"
    },
    "chain_id": {
      "description": "Chain identifier returned by create_chain",
      "title": "Chain Id",
      "type": "string"
    },
    "stage": {
      "description": "Stage name to submit for (must be the chain's current stage)",
      "title": "Stage",
      "type": "string"
    },
    "structured_data": {
      "additionalProperties": true,
      "description": "The document's extracted fields as key/value pairs. Keys are open by design - your Blueprint's rules define what is checked",
      "title": "Structured Data",
      "type": "object"
    }
  },
  "required": [
    "api_key",
    "chain_id",
    "stage",
    "structured_data"
  ],
  "title": "submit_chain_stageArguments"
}

Audit a handoff between two chain stages: a context capsule of verified facts from the prior stage, and (if proposed_data is given) a compatibility verdict that catches fields mutated in transit. Siblings: create_chain, submit_chain_stage.

{
  "type": "object",
  "properties": {
    "api_key": {
      "description": "GeodesicAI API key (gai_...)",
      "title": "Api Key",
      "type": "string"
    },
    "chain_id": {
      "description": "Chain identifier returned by create_chain",
      "title": "Chain Id",
      "type": "string"
    },
    "from_stage": {
      "description": "Completed stage name (agent A)",
      "title": "From Stage",
      "type": "string"
    },
    "to_stage": {
      "description": "Stage about to start (agent B)",
      "title": "To Stage",
      "type": "string"
    },
    "proposed_data": {
      "additionalProperties": true,
      "default": null,
      "description": "Data agent B intends to submit; checked for mutation against agent A's verified fields",
      "title": "Proposed Data",
      "type": "object"
    }
  },
  "required": [
    "api_key",
    "chain_id",
    "from_stage",
    "to_stage"
  ],
  "title": "handoff_auditArguments"
}

Promote a rule discovered by discover_patterns into Blueprint-ready form.

{
  "type": "object",
  "properties": {
    "api_key": {
      "description": "GeodesicAI API key (gai_...)",
      "title": "Api Key",
      "type": "string"
    },
    "rule_id": {
      "description": "Discovered rule ID from discover_patterns",
      "title": "Rule Id",
      "type": "string"
    },
    "blueprint": {
      "default": "default",
      "description": "Blueprint name (workflow_name) to use",
      "title": "Blueprint",
      "type": "string"
    }
  },
  "required": [
    "api_key",
    "rule_id"
  ],
  "title": "approve_ruleArguments"
}

Reject a discovered candidate rule so it will not be promoted into a Blueprint. Pair with approve_rule after discover_patterns.

{
  "type": "object",
  "properties": {
    "api_key": {
      "description": "GeodesicAI API key (gai_...)",
      "title": "Api Key",
      "type": "string"
    },
    "rule_id": {
      "description": "Discovered rule ID from discover_patterns",
      "title": "Rule Id",
      "type": "string"
    },
    "blueprint": {
      "default": "default",
      "description": "Blueprint name (workflow_name) to use",
      "title": "Blueprint",
      "type": "string"
    }
  },
  "required": [
    "api_key",
    "rule_id"
  ],
  "title": "reject_ruleArguments"
}

Retrieve the document categories a discover_patterns session identified (counts, distinguishing fields, domain hints). Read-only; returns status=no_session if discovery has not run for this namespace.

{
  "type": "object",
  "properties": {
    "api_key": {
      "description": "GeodesicAI API key (gai_...)",
      "title": "Api Key",
      "type": "string"
    },
    "blueprint": {
      "default": "default",
      "description": "Blueprint name (workflow_name) to use",
      "title": "Blueprint",
      "type": "string"
    }
  },
  "required": [
    "api_key"
  ],
  "title": "structural_typesArguments"
}

Split the error between original and corrected values into direct rule violations, boundary violations, and systemic structural error, with per-field contributions. Use with a known-correct version to diff against; use analyze_anomaly when you only have the suspicious payload. Diagnostics-tier tool.

{
  "type": "object",
  "properties": {
    "api_key": {
      "description": "GeodesicAI API key (gai_...)",
      "title": "Api Key",
      "type": "string"
    },
    "original_values": {
      "additionalProperties": true,
      "description": "Original numeric field values as {field: number}",
      "title": "Original Values",
      "type": "object"
    },
    "corrected_values": {
      "additionalProperties": true,
      "description": "Corrected/expected numeric field values as {field: number}",
      "title": "Corrected Values",
      "type": "object"
    },
    "derivation_rules": {
      "default": null,
      "description": "Math rules as objects. Types: add, subtract, multiply, divide, round, copy, sum (multi-operand), items_multiply, items_sum. Each needs 'type' plus its fields; see the blueprint_guide prompt",
      "items": {},
      "title": "Derivation Rules",
      "type": "array"
    },
    "formal_constraints": {
      "default": null,
      "description": "Constraint objects. Types incl. magnitude_anchor {field,min,max}, relative_anchor {field,reference_field,ratio_min,ratio_max}, max_action_threshold {field,threshold,on_violation}, required_fields {fields}, equals, range, in_set, regex_match, items_magnitude_anchor; see the blueprint_guide prompt",
      "items": {},
      "title": "Formal Constraints",
      "type": "array"
    },
    "blueprint": {
      "default": null,
      "description": "Load rules from this Blueprint instead of passing them inline",
      "title": "Blueprint",
      "type": "string"
    }
  },
  "required": [
    "api_key",
    "original_values",
    "corrected_values"
  ],
  "title": "decompose_failureArguments"
}

Summarize an already-computed state_vector into a confidence level (high/medium/low) with a recommendation. Post-hoc digest - use analyze_anomaly or check_drift for fresh analysis of raw data.

{
  "type": "object",
  "properties": {
    "api_key": {
      "description": "GeodesicAI API key (gai_...)",
      "title": "Api Key",
      "type": "string"
    },
    "state_vector": {
      "additionalProperties": true,
      "description": "state_vector object from a prior validate or get_execution_trace result",
      "title": "State Vector",
      "type": "object"
    }
  },
  "required": [
    "api_key",
    "state_vector"
  ],
  "title": "geometric_confidenceArguments"
}

Structural realization analysis of a payload against the Blueprint's reference configuration (requires a 'realization' block; otherwise status=skipped). Diagnostics-tier tool; prefer validate or analyze_anomaly for standard checks.

{
  "type": "object",
  "properties": {
    "api_key": {
      "description": "GeodesicAI API key (gai_...)",
      "title": "Api Key",
      "type": "string"
    },
    "structured_data": {
      "additionalProperties": true,
      "description": "The document's extracted fields as key/value pairs. Keys are open by design - your Blueprint's rules define what is checked",
      "title": "Structured Data",
      "type": "object"
    },
    "blueprint": {
      "default": "default",
      "description": "Blueprint name (workflow_name) to use",
      "title": "Blueprint",
      "type": "string"
    }
  },
  "required": [
    "api_key",
    "structured_data"
  ],
  "title": "check_realizationArguments"
}

Check whether recent submissions still match the established pattern for this Blueprint. Returns a stability verdict and observation count.

{
  "type": "object",
  "properties": {
    "api_key": {
      "description": "GeodesicAI API key (gai_...)",
      "title": "Api Key",
      "type": "string"
    },
    "structured_data": {
      "additionalProperties": true,
      "description": "The document's extracted fields as key/value pairs. Keys are open by design - your Blueprint's rules define what is checked",
      "title": "Structured Data",
      "type": "object"
    },
    "blueprint": {
      "default": "default",
      "description": "Blueprint name (workflow_name) to use",
      "title": "Blueprint",
      "type": "string"
    }
  },
  "required": [
    "api_key",
    "structured_data"
  ],
  "title": "check_driftArguments"
}

Go/no-go for a real-world action (payment, filing, API write): runs full validation, then the Blueprint's execution gate. authorized=true only on PASS; REVIEW means do not proceed automatically. Different from validate: validate asks is this data correct, authorize_execution asks should this action happen.

{
  "type": "object",
  "properties": {
    "api_key": {
      "description": "GeodesicAI API key (gai_...)",
      "title": "Api Key",
      "type": "string"
    },
    "structured_data": {
      "additionalProperties": true,
      "description": "The document's extracted fields as key/value pairs. Keys are open by design - your Blueprint's rules define what is checked",
      "title": "Structured Data",
      "type": "object"
    },
    "blueprint": {
      "description": "Blueprint name (workflow_name) to use",
      "title": "Blueprint",
      "type": "string"
    }
  },
  "required": [
    "api_key",
    "structured_data",
    "blueprint"
  ],
  "title": "authorize_executionArguments"
}

Load a prebuilt Blueprint template (invoices, timecards, legal, POs, claims). Call without pack_id to list packs; then create_blueprint to save a customized copy.

{
  "type": "object",
  "properties": {
    "api_key": {
      "description": "GeodesicAI API key (gai_...)",
      "title": "Api Key",
      "type": "string"
    },
    "pack_id": {
      "default": null,
      "description": "Rule pack ID; omit to list available packs",
      "title": "Pack Id",
      "type": "string"
    }
  },
  "required": [
    "api_key"
  ],
  "title": "load_rule_packArguments"
}

Run validation and return the per-node execution trace (node names, deterministic flags, timing) plus the verdict and determinism hash. Use validate for normal operation; this is for debugging and audit preparation.

{
  "type": "object",
  "properties": {
    "api_key": {
      "description": "GeodesicAI API key (gai_...)",
      "title": "Api Key",
      "type": "string"
    },
    "structured_data": {
      "additionalProperties": true,
      "description": "The document's extracted fields as key/value pairs. Keys are open by design - your Blueprint's rules define what is checked",
      "title": "Structured Data",
      "type": "object"
    },
    "blueprint": {
      "default": "default",
      "description": "Blueprint name (workflow_name) to use",
      "title": "Blueprint",
      "type": "string"
    }
  },
  "required": [
    "api_key",
    "structured_data"
  ],
  "title": "get_execution_traceArguments"
}

Check replay commitment integrity and compare recorded execution components. Version 4 includes reference context and the final result/status. A match compares commitments; this tool does not rerun the workflow or reconstruct historical reference populations, and does not prove factual correctness.

{
  "type": "object",
  "properties": {
    "api_key": {
      "description": "GeodesicAI API key (gai_...)",
      "title": "Api Key",
      "type": "string"
    },
    "contract_a": {
      "additionalProperties": true,
      "description": "replay_contract object from one execution",
      "title": "Contract A",
      "type": "object"
    },
    "contract_b": {
      "additionalProperties": true,
      "description": "replay_contract object to compare against contract_a",
      "title": "Contract B",
      "type": "object"
    }
  },
  "required": [
    "api_key",
    "contract_a",
    "contract_b"
  ],
  "title": "verify_replayArguments"
}

This account's plan, key usage, Blueprint counts, and the deployed platform build fingerprint (version, build, deployed).

{
  "type": "object",
  "properties": {
    "api_key": {
      "description": "GeodesicAI API key (gai_...)",
      "title": "Api Key",
      "type": "string"
    }
  },
  "required": [
    "api_key"
  ],
  "title": "account_statusArguments"
}

Permanently delete a Blueprint and revoke its API keys. Irreversible; requires confirm=true. Account-level keys are unaffected.

{
  "type": "object",
  "properties": {
    "api_key": {
      "description": "GeodesicAI API key (gai_...)",
      "title": "Api Key",
      "type": "string"
    },
    "workflow_name": {
      "description": "Blueprint to delete; its API keys are revoked",
      "title": "Workflow Name",
      "type": "string"
    },
    "confirm": {
      "default": false,
      "description": "Must be true to confirm this irreversible action",
      "title": "Confirm",
      "type": "boolean"
    }
  },
  "required": [
    "api_key",
    "workflow_name"
  ],
  "title": "delete_blueprintArguments"
}

Update an existing Blueprint in place. Only passed fields change; pass [] to clear a list. workflow_name cannot be renamed and existing API keys keep working. Different from create_blueprint: modifies an existing Blueprint, mints no new key.

{
  "type": "object",
  "properties": {
    "api_key": {
      "description": "GeodesicAI API key (gai_...)",
      "title": "Api Key",
      "type": "string"
    },
    "workflow_name": {
      "description": "Unique Blueprint identifier; the value passed as 'blueprint' in validate",
      "title": "Workflow Name",
      "type": "string"
    },
    "customer_name": {
      "default": null,
      "description": "Organization or project name (also used for storage folder naming)",
      "title": "Customer Name",
      "type": "string"
    },
    "mode": {
      "default": null,
      "description": "New mode: observe or enforce; omit to keep current",
      "title": "Mode",
      "type": "string"
    },
    "extracted_fields": {
      "default": null,
      "description": "Field names the agent extracts from source data, e.g. ['vendor','qty','unit_cost']",
      "items": {},
      "title": "Extracted Fields",
      "type": "array"
    },
    "derived_fields": {
      "default": null,
      "description": "Field names the platform computes from other fields, e.g. ['subtotal','total']",
      "items": {},
      "title": "Derived Fields",
      "type": "array"
    },
    "derivation_rules": {
      "default": null,
      "description": "Math rules as objects. Types: add, subtract, multiply, divide, round, copy, sum (multi-operand), items_multiply, items_sum. Each needs 'type' plus its fields; see the blueprint_guide prompt",
      "items": {},
      "title": "Derivation Rules",
      "type": "array"
    },
    "formal_constraints": {
      "default": null,
      "description": "Constraint objects. Types incl. magnitude_anchor {field,min,max}, relative_anchor {field,reference_field,ratio_min,ratio_max}, max_action_threshold {field,threshold,on_violation}, required_fields {fields}, equals, range, in_set, regex_match, items_magnitude_anchor; see the blueprint_guide prompt",
      "items": {},
      "title": "Formal Constraints",
      "type": "array"
    },
    "semantic_checks": {
      "default": null,
      "description": "Domain-specific semantic check objects",
      "items": {},
      "title": "Semantic Checks",
      "type": "array"
    },
    "require_math": {
      "default": null,
      "description": "Validate mathematical relationships",
      "title": "Require Math",
      "type": "boolean"
    },
    "require_consistency": {
      "default": null,
      "description": "Check internal field consistency",
      "title": "Require Consistency",
      "type": "boolean"
    },
    "require_coherence": {
      "default": null,
      "description": "Check cross-field plausibility",
      "title": "Require Coherence",
      "type": "boolean"
    },
    "require_provenance": {
      "default": null,
      "description": "Require extraction source locations for fields",
      "title": "Require Provenance",
      "type": "boolean"
    },
    "require_high_assurance": {
      "default": null,
      "description": "Strictest mode: every check must pass",
      "title": "Require High Assurance",
      "type": "boolean"
    },
    "enable_anomaly_detection": {
      "default": null,
      "description": "Flag records that break no rules but do not fit the reference pattern",
      "title": "Enable Anomaly Detection",
      "type": "boolean"
    },
    "enable_drift_tracking": {
      "default": null,
      "description": "Track pattern stability across batches",
      "title": "Enable Drift Tracking",
      "type": "boolean"
    }
  },
  "required": [
    "api_key",
    "workflow_name"
  ],
  "title": "update_blueprintArguments"
}

List this account's API keys (masked) with their Blueprint bindings.

{
  "type": "object",
  "properties": {
    "api_key": {
      "description": "GeodesicAI API key (gai_...)",
      "title": "Api Key",
      "type": "string"
    }
  },
  "required": [
    "api_key"
  ],
  "title": "list_api_keysArguments"
}

Replace an API key with a fresh one. The old key stops working immediately; the new key inherits its bindings.

{
  "type": "object",
  "properties": {
    "api_key": {
      "description": "GeodesicAI API key (gai_...)",
      "title": "Api Key",
      "type": "string"
    },
    "key_to_rotate": {
      "description": "The gai_ key to rotate; it stops working immediately",
      "title": "Key To Rotate",
      "type": "string"
    }
  },
  "required": [
    "api_key",
    "key_to_rotate"
  ],
  "title": "rotate_api_keyArguments"
}

Permanently delete one of the caller's API keys. DESTRUCTIVE — agents using the deleted key will receive auth errors immediately. The Blueprint a key was tied to (if any) is NOT affected; only the credential is revoked. To delete a Blueprint and all its keys, use delete_blueprint. The target key can be specified two ways: - As the full key string (gai_...). - As a key_id (SHA-256 hash from list_api_keys).

{
  "type": "object",
  "properties": {
    "api_key": {
      "description": "GeodesicAI API key (gai_...)",
      "title": "Api Key",
      "type": "string"
    },
    "key_to_delete": {
      "description": "The gai_ key to delete",
      "title": "Key To Delete",
      "type": "string"
    },
    "confirm": {
      "default": false,
      "description": "Must be true to confirm this irreversible action",
      "title": "Confirm",
      "type": "boolean"
    }
  },
  "required": [
    "api_key",
    "key_to_delete"
  ],
  "title": "delete_api_keyArguments"
}