🤓 Latest Submissions
Governed Agents
AI agents are being deployed into enterprise workflows right now. But they operate with no governance layer. They pass data to each other blindly. Nobody verifies the handoff happened correctly. Nobody checks that outputs follow organizational rules. Nobody confirms which agent is actually sending the data. There is no audit trail. We built the governance infrastructure that sits between agents. Not just the agents themselves, the layer that makes agents trustworthy enough to deploy in regulated environments. Four governance primitives: Structure — Iron-Thread validates every agent output against a registered schema and builds a SHA-256 hash chain. Handoff — ChainThread wraps every agent-to-agent handoff in a signed envelope, verifies sender identity, and checks the payload against a contract. Compliance — PolicyThread evaluates every output against organizational rules using keyword checks and semantic evaluation via Claude. Identity — AgentID issues cryptographic credentials to every agent and makes trust decisions via reputation scores. Demonstrated on a loan review pipeline running through Band. Four agents : Intake, Risk, Compliance, Decision coordinate in real time. Every handoff is verified. Every output is validated. Every compliance rule is checked. A tamper-evident audit trail runs from first input to final decision, cryptographically signed and provable to a regulator. Built with Band for coordination and Thread Suite for governance. The governed multi-agent platform for regulated industries.
19 Jun 2026
Org Intelligence OS
Org Intelligence OS is a 5-layer AI platform that sits at the center of any organization, ingests everything happening inside and around it, and delivers structured intelligence that tells decision-makers not just what is happening, but what to do and why. Layer 1 — Data Foundation: Ingests PDF, DOCX, and TXT documents. Automatically scores each source across 4 bias dimensions, epistemic, institutional, cultural, and temporal, and flags when the source pool is dangerously one-sided. Layer 2 — Pattern Intelligence Core: Extracts structured knowledge claims from every document using Claude API. Detects contradictions across sources (Type I through IV with 7 named subtypes) and synergies (5 types). Selects a synthesis pattern, Wedge, Dialectic, Mosaic, or Catalyst, and generates a full intelligence brief with provenance tracking and confidence scoring. Layer 3 — Simulation Chamber: Takes the intelligence brief as a seed document. Spawns AI agent-personas representing the relevant stakeholder universe, investors, regulators, NGOs, competitors, analysts. Runs the decision forward. Returns a verdict with confidence score, dominant reaction, critical vulnerabilities, and named early warning signals. Layer 4 — Prescription Engine: Ranks every recommendation across 7 dimensions — confidence, financial impact, stakeholder resistance, time to first signal, simulation alignment, implementation complexity, and sequence position. Integrates Layer 3 simulation confidence into an overall prescription score. Layer 5 — Strategic Command Center: A live dashboard where analysts manage projects, upload documents, trigger the pipeline, read intelligence briefs, launch simulations, and review ranked prescriptions. Validated on real Holcim Group documents, 4 sources, 90 knowledge nodes, 14 contradictions detected, 50 synergies mapped, 0.829 overall prescription confidence.
19 May 2026
AgentRx: One-Click Reliability Audit for AI Agents
AgentRx is a one-click reliability audit tool for AI agents, built during the IBM Bob Hackathon with IBM Bob as the AI development partner. The problem: developers ship AI agents to production without knowing if their outputs are reliable, their behavior is consistent, or their responses comply with organizational rules. They find out when something breaks. AgentRx solves this with three automated checks powered by the Thread Suite — nine open-source AI agent reliability tools built by Eugene Dayne Mawuli (BiteLance, Accra, Ghana): Structure Check (Iron-Thread): Validates that the agent returns well-formed, consistent output using JSON schema validation. Catches malformed responses before they reach a database. Behavior Check (TestThread): Runs three automated behavioral test cases against the live agent endpoint basic response, instruction following, and simple arithmetic. Measures pass rate and latency. Compliance Check (PolicyThread): Evaluates the agent's responses against domain-specific compliance policies for General, Medical, Finance, and Legal use cases. Catches harmful content, specific medical diagnoses, investment advice guarantees, and legal outcome promises. IBM Bob was used throughout the build to read the Thread Suite production codebases, design the integration architecture, and implement robust retry logic with exponential backoff for handling Render free tier cold starts. AgentRx returns a Reliability Score from 0 to 100 with specific failures and actionable recommendations for each check.
17 May 2026
