Codium

Codium AI is at the forefront of transforming the way developers review pull requests with its suite of AI-powered tools. By leveraging advanced AI algorithms, Codium AI aims to streamline the pull request review process, providing developers with automated analysis, feedback, suggestions, and more. This technology empowers developers to code smarter, create more value, and boost confidence when pushing their code changes.

General
Author	Codium AI
Repository	https://github.com/Codium-ai/pr-agent
Type	AI-Powered Code Review abd Testing Tool

Start building with Codium AI

Visit the GitHub repository to explore and access all the Codium AI resources. Here, you can find code samples, documentation, and more to help you integrate Codium AI into your development workflow.

Codium Tutorials

Codium AI Resourses

Explore a wealth of Codium AI resources to enhance your pull request review process. These resources provide a great head start when revolutionizing your code review workflow.

AuditPro Agent

The Problem Traditional financial audits and risk assessments are severely bogged down by massive volumes of unstructured documents, including general ledgers, invoices, contracts, and regulatory filings. Auditors spend countless hours manually parsing text, tracking down transactional anomalies, and cross-checking compliance policies. This manual overload creates severe bottlenecks, raises the risk of human error, and delays critical business insights. The Solution Our project introduces an intelligent, autonomous financial and audit agent built for the AMD Hackathon. By integrating Evolus AI, the platform seamlessly handles heavy document chunking, semantic embedding, and high-performance text indexing to instantly process complex ledgers. Running natively on robust AMD infrastructure (leveraging AMD Instinct GPUs and the ROCm software stack), the agent executes automated anomaly detection, flags fiscal risks, verifies transaction compliance, and delivers real-time audit results directly through an intuitive dashboard—drastically cutting down audit cycle times and streamlining compliance workflows.

ryzen ai

i have build a project to amazing arms from to use ai and integrate ml or apply rienforcement algo i have build a project to amazing arms from to use ai and integrate ml or apply rienforcement algo i have build a project to amazing arms from to use ai and integrate ml or apply rienforcement algo i have build a project to amazing arms from to use ai and integrate ml or apply rienforcement algo i have build a project to amazing arms from to use ai and integrate ml or apply rienforcement algo i have build a project to amazing arms from to use ai and integrate ml or apply rienforcement algo i have build a project to amazing arms from to use ai and integrate ml or apply rienforcement algo i have build a project to amazing arms from to use ai and integrate ml or apply rienforcement algo

Fresno LUX: A Living City

Fresno LUX is a prototype for a living intelligence layer across an entire city. Fresno already contains enormous amounts of intelligence: residents know where problems exist, nonprofits know where help is needed, businesses know where opportunities exist, educators understand gaps in learning, researchers generate knowledge, and government maintains critical infrastructure and data. The problem is that much of this intelligence is fragmented across disconnected systems. Fresno LUX connects those pieces. The platform creates a shared city graph where people, organizations, resources, needs, projects, observations, experiments, and outcomes become connected nodes. Instead of simply collecting data, LUX continuously asks: What is happening? What do we know? What is missing? What can we test? What changed? A community problem can become an observable signal. That signal can become an experiment. The experiment produces evidence. Evidence changes the system's understanding. That updated understanding informs the next action. Observe → Connect → Understand → Act → Measure → Learn → Converge. The goal is to move Fresno from disconnected information toward collective intelligence—a system where knowledge can travel across boundaries and successful solutions can be discovered, tested, measured, and replicated. LUX is designed to make the invisible connections visible: the relationship between a neighborhood and its resources, between a student and educational opportunity, between a business and workforce capacity, between environmental conditions and community health, and between an idea and measurable real-world impact. This hackathon prototype demonstrates the beginning of a Living Fresno Graph: a city that can observe itself, learn from evidence, and coordinate toward better outcomes.

vis-Neuclea: Scoped Agent Infrastructure for APIs

THE WEB WASN'T BUILT FOR AGENTS Modern web architecture was engineered for humans. Forcing AI agents into it creates three bottlenecks: bloated ReAct loops that burn thousands of tokens per query, credentials that have to live on third-party backend servers, and bespoke API wrappers that slow every new integration. THE SOLUTION: VIS + NEUCLEA VisNeucla turns any OpenAPI 3.x spec into a conversational AI agent in minutes — no boilerplate, no glue code. Vis is a TypeScript CLI that parses an OpenAPI spec through a strict one-way pipeline (core → parser → templates → generator → cli) and emits a typed, production-ready MCP server: Zod-validated tools, automatic parameter routing across query/path/header/body, clean JSON Schema for tool discovery, and an optional containerized observability stack (Prometheus, Loki, Tempo, Grafana) for metrics, logs, and tracing. Neuclea is a Go WebSocket gateway that runs a round-based ReAct agent (max 3 rounds) against that MCP server. Independent tools in a round execute in parallel via goroutines. Every design choice is tuned for tokens: first-sentence-only tool descriptions, required-params-only listings, name-only history summaries, and capped output at each planning and formatting step. A Markov-chain predictor learns tool-call sequences to power autocomplete, and the MCP client pool handles per-endpoint rate limiting, retries with backoff, and cold-start detection. THE RESULT A typical two-round query costs ~1,800–2,200 tokens — roughly 5x less than an unconstrained agent loop — while streaming live reasoning steps over WebSocket for a transparent, responsive console. TECH STACK Go 1.25, TypeScript, gorilla/websocket, Zod, pnpm workspaces, Fireworks AI (MiniMax M2P7 / GLM series), Ollama, Model Context Protocol, OpenAPI 3.x, Prometheus, Grafana, Loki, Tempo, Promtail.

CivicOps Command Platform

CivicOps Command Platform is a multi-agent civic emergency dispatch system built for the Band of Agents hackathon. It helps municipalities turn scattered citizen reports into structured, trackable response workflows. When a resident reports an issue such as a burst water pipe, outage, blocked road, waste problem, or public safety concern, CivicOps moves the incident through a coordinated agent chain. The Intake Agent captures the location, category, urgency, and description. The Triage Agent classifies the incident and determines priority. The Dispatch Agent routes it to the correct municipal response team. The Public Communications Agent prepares a citizen-friendly status update. The Audit and Supervisor Agent records the decision trail for accountability. Band is used as the coordination layer for the multi-agent workflow, showing how focused agents can hand off structured context instead of relying on one generic chatbot. CivicOps also includes optional provider readiness for AI/ML API and Featherless AI, while keeping deterministic fallback mode available so the demo remains reliable even when external providers are not available. Live demo: https://culltron.app/hackathon Demo video: https://youtu.be/mlmoPMdbyzg?feature=shared Pitch deck: https://github.com/AeriesAgi/civicops/blob/main/CivicOps_Band_of_Agents_Pitch_Deck_QA.pdf GitHub repo: https://github.com/AeriesAgi/civicops The deployed platform includes a judge-facing hackathon page, Band demo room, mobile preview, app/download area, technology partner section, and a full burst-water-pipe scenario showing the end-to-end civic response loop.

BOM Squad: Hardware Procurement Agent Crew

Designing hardware means losing days to sourcing — hunting parts across stores, checking what fits, and balancing a budget by hand. BOM Squad fixes that. You need 10 parts. And mostly none of them are on the same website. Takes a lot of grind, and luck and research, with no guranteee that u wil even ever find it. BOM saves exactly this whole time for you. You describe a build in plain English ("a CoreXY 3D printer, ₹10,000 — I need a 32-bit mainboard, steppers, a hotend, and a PSU"). A crew of specialist agents, coordinating live through Band, takes it from there: • Intake structures your request into clear requirements. • Sourcing searches the live web for real, in-stock parts and prices — and recruits teammates into the room when needed. • A Specialist validates the build-defining component against the rest of the build. • Trust scores each seller for risk. • Budget totals the cart in code; if you're over budget it re-optimizes — swapping the priciest part for a cheaper real alternative until it fits — then hands the decision back to you. • A Replacement Scout is recruited only when a part can't be sourced. • With a lot other re-optimiziations. You approve the premium cart or the budget-fit one, and the order is saved with links to buy every part. The agent does the fetching, Maths to the code, and decision for you. Giving you near Zero Hallucination risk. (occurs cause of AI, sometimes.) Built solo for the BandHack by a mechatronics hobbyist who lived the problem.