ai.geodesiclabs/governance-platform
ai.geodesiclabs/governance-platformDeterministic AI governance platform. Validates agent outputs, discovers patterns, solves math.
Tools · 38
Validate structured data against a Blueprint's rules. Returns PASS, FAIL, or REVIEW. The platform checks mathematical accuracy (do the numbers add up?), structural consistency (do the fi…
Validate structured data and automatically compute repairs if it fails. Single call that combines validate + repair. Different from validate: validate returns only the verdict; if the …
Create a Blueprint — a governance contract that defines validation rules. A Blueprint tells the platform what "correct" means for your data: which fields exist, what math must hold betwe…
List Blueprints owned by the calling account. Returns each Blueprint's name, workflow identifier, mode, and field/rule/constraint counts. Use the workflow_name as the 'blueprint' par…
Compute the nearest valid point on the constraint manifold. Given structured data with errors, missing values, or inconsistencies, returns field-by-field repair suggestions with geometri…
Check whether a set of constraints can be simultaneously satisfied. Detects fundamental conflicts in the constraint system that cannot be resolved by adjusting any single field. Returns …
Certify the structural health of a Blueprint BEFORE deployment. Analyzes how much tolerance the Blueprint's rules and constraints leave between a passing and a failing document. Returns …
Compare two structured payloads at both invariance levels of the platform's dual-hash design. content_hash — invariant to field ORDER and numeric FORMATTING (5 vs 5.…
Quality-govern an in-progress AI generation step BEFORE its output is used. Complements validate (which checks finished structured documents): this checks the generation PROCESS. Tas…
Retrieve the durable audit trail for a governed generation: every recorded check (action, confidence, hashes, runtime) and the full step-by-step event log. Args: api_key: Geo…
Recent governance decisions across all generations — a quick health view of what automated quality control has been approving, holding, and escalating. Args: api_key: Geodesi…
Independently re-verify a validation certificate. Certificates are issued with every governed execution (the `certificate` field of the output contract). They are self-contained: thi…
Measure how brittle a Blueprint's verdict is to changes in its constraint bounds. Sweeps the numeric constraint windows tighter and wider and reports: the stable band around the curr…
Deterministic forward reasoning — show what valid states are reachable. Given the current data state, generates candidate next states by applying Blueprint rules, then ranks them by stru…
Feed a batch of structured data to discover patterns deterministically. No Blueprint required — discovers rules from the data itself. Analyzes the batch using motif discovery, structural…
Find the shortest path from an invalid state to a valid one. Given data that fails validation, computes a sequence of minimal field changes that would bring the data into compliance with…
Compare outcomes under different rule sets. Given the same data, runs trajectory analysis under two different sets of rules/constraints and shows how the valid state space differs. U…
Deep anomaly analysis with geometric proof. No Blueprint required. Explains WHY data is anomalous using three independent methods: 1. Structural fingerprinting — distance from the learne…
Create a multi-agent sequential execution chain. Defines a pipeline where multiple agents process data in sequence. Each stage is validated against the Blueprint before the next stage …
Submit data for a chain stage. The platform validates the data using the chain's Blueprint, then advances the chain if validation passes. The response includes the next stage info and an…
Audit a handoff between two chain stages. Returns a context capsule with verified facts from the prior stage and checks structural compatibility of proposed data for the next stage. …
Promote a discovered rule into Blueprint-compatible format. After running discover_patterns, use this to approve high-confidence rules for inclusion in a Blueprint. Args: ap…
Reject a discovered candidate rule so it will not be promoted into a Blueprint. Use after running discover_patterns when reviewing the candidate rules it produced. Pair with approve_rule…
Get auto-discovered structural type classifications from a discovery session. After running discover_patterns, returns the structural categories the platform identified in the data — wit…
Hodge-style decomposition of validation failures. Splits the error between original and corrected values into three orthogonal components: - Exact: direct rule violations (a field br…
Compute a composite geometric confidence score from validation signals. No Blueprint required — works on any validate result. Combines six weighted signals into a single confidence score…
Run structural realization analysis on a payload. Embeds the payload via the Blueprint's declared embedding schema, projects it onto the Blueprint's reference subspace, and returns a…
Check whether the data pattern has shifted since previous observations. Works with or without a Blueprint. Monitors structural stability across a stream of data. Detects regime changes w…
Decide whether an action should be allowed to proceed. Runs full validation, then applies the Blueprint's execution gate. Returns a simple allow/block decision with reasoning. Use t…
Load a prebuilt Blueprint template for fast onboarding. Rule Packs are ready-made governance configurations for common use cases. Call with no pack_id to list all available packs. Call w…
Run validation and return the detailed execution trace. Shows the exact sequence of validation nodes that ran, whether each was deterministic, and the runtime of each node. Use for d…
Verify that two execution replay contracts represent the same deterministic result. This is the programmatic proof of GeodesicAI's core promise: same input + same rules = same result, ev…
Report the calling account's plan, key usage, and limits. Use this to introspect what the caller is allowed to do. Agents that hit rate limits or key-count caps can call this to explain …
Permanently delete a Blueprint and all of its API keys. DESTRUCTIVE — cannot be undone. Cascading effects: - The Blueprint's template_config.json is removed from disk. - All Blueprin…
Update an existing Blueprint's configuration in place. Only fields you pass are updated; fields you omit keep their current values. To clear a list field (e.g. remove all rules), pas…
List all API keys owned by the calling account. Returns a masked representation of each key plus a stable key_id (SHA-256 hash) that can be used with rotate_api_key and delete_api_ke…
Rotate one of the caller's API keys. The old key stops working immediately; a new key with the same scope is issued. The new key inherits the old key's scope: - An account-level ke…
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 affe…
How to use
Add to your Claude Desktop / Cursor / Cline MCP config:
{
"mcpServers": {
"ai.geodesiclabs/governance-platform": {
"url": "https://app.geodesiclabs.ai/mcp",
"transport": "streamable-http"
}
}
}