Search Results for "evidence"
Sort By:
Showing 175 open source projects for "evidence"
View related business solutions-
Build Agents and Models on One PlatformGemini Enterprise Agent Platform is Google Cloud's comprehensive platform for developers to build, scale, govern, and optimize agents and models. Choose from Google's most advanced models and third-party models like Anthropic's Claude Model Family.
-
Host LLMs in Production With On-Demand GPUsDeploy your model, get an endpoint, pay only for compute time. No GPU provisioning or infrastructure management required.
-
1
HowToLiveBetter
Evidence-based living guidelines sorted by cost-effectiveness
HowToLiveBetter is an evidence-focused interactive life guide that organizes hundreds of practical recommendations by cost, expected benefit, and strength of evidence. It covers health, longevity, emergency response, personal finance, legal risks, unemployment, entrepreneurship, relationships, parenting, travel, and skill development. Each recommendation explains what resources it costs and what outcome it is intended to improve. -
2
CRM
An open-source, agentic-first CRM
...The agent operates on its own schedule and work queue, investigates contacts and companies, records supported facts, and schedules future checks. Rather than guessing personal details, it prices evidence and writes strong observations while sending weak evidence to a human for resolution. The interface manages contacts, companies, deals, tasks, pipeline views, and agent activity. Email threads, meetings, signature blocks, optional web research, and enrichment services can provide evidence. Durable sessions survive reloads and deployments, while a restricted sandbox prevents direct database access and network egress. ... -
3
PRAXIST
Autonomous research system for measurable computer-executable research
...It turns an already runnable project into a persistent research process instead of a series of disconnected prompts. Parallel research peers explore competing hypotheses and implementations while evaluators convert outcomes into structured evidence. That evidence is carried across generations so later work can build on promising strategies and avoid repeating weak ones. The system supports multi-metric evaluation, quality-diversity methods, resource-aware scheduling, and durable evidence tracking. Runs can be monitored, stopped, resumed, and replayed, with the task project remaining the source of truth for objectives, metrics, baselines, and domain constraints. -
4
Better Harness
Help your coding agents get better at getting better
...Instead of reviewing only the final code change, it examines how an agent understood the task, executed work, validated results, delivered safely, and captured reusable lessons. It gathers project and session evidence while marking missing or unobserved behavior explicitly. Reports organize findings across five Agent Work Loop dimensions and connect each conclusion to visible supporting evidence. Supported gaps become prioritized recommendations with expected outcomes, repair boundaries, and acceptance checks. Repeated reports can be compared through a history view to observe workflow trends without claiming unsupported causation. ... -
MongoDB Atlas runs apps anywhereMongoDB Atlas gives you the freedom to build and run modern applications anywhere—across AWS, Azure, and Google Cloud. With global availability in over 115 regions, Atlas lets you deploy close to your users, meet compliance needs, and scale with confidence across any geography.
-
5
how-to-live-better
Online single-page reading version of the 528-item "High-Value Life"
How to Live Better is a self-contained web reader for the open-source High Cost-Performance Life Guide. It renders all 32 sections and 528 recommendations from the upstream project into a single searchable HTML page. Each recommendation is presented as a card with evidence grade, value tier, and cost labels. Users can search the full guide, filter for top-grade evidence, or show only simplified explanations. A synchronized table of contents, responsive mobile layout, dark mode, and print support improve navigation. The page uses no external resources, so it can be opened and read entirely offline. ... -
6
Harness-1
Ultra Recipe for Training Long-Horizon Search Agents
Harness-1 is a 20B search agent trained with reinforcement learning inside a stateful retrieval harness. It is designed for long-horizon search tasks where the model must search, inspect documents, curate evidence, verify claims, and decide when enough evidence has been gathered. The harness externalizes search state, including candidate documents, evidence links, verification records, and budget-aware context. This lets the policy focus on higher-level decisions instead of trying to keep every detail inside the model context. The repository includes inference utilities, training scripts, evaluation runners, dataset tools, and documentation for running the released checkpoint. ... -
7
Gajae-Code
Gajae Code MVP
...It runs beside existing tools rather than hiding inside a specific agent runtime, so users can apply it to a chosen repository or isolated worktree. The workflow is built around clarifying requirements, planning before mutation, turning approved plans into goals, and tracking evidence until completion. It includes tmux-backed execution options for coordinating parallel workers when a task is large enough to benefit from them. Gajae-Code is useful for developers who want coding agents to behave less like single-prompt assistants and more like disciplined project collaborators. Its main value is giving AI coding work a clearer loop for interviews, plans, goals, execution checks, and evidence. -
8
HOL Guard
Open-source antivirus for AI agents
...It integrates with major coding agents including Claude Code, Codex, Cursor, Gemini CLI, OpenCode, and GitHub Copilot CLI. Local operation requires no cloud account, while optional cloud services add synchronized evidence, shared policies, fleet visibility, and team approvals. -
9
OpenResearch
Turn your coding agents into research agents
...Agents can review literature, propose hypotheses, modify code, launch experiments, inspect results, and generate research artifacts. Experiments are tracked through a Git-native tree that preserves lineage and immutable snapshots. Logs, diffs, files, results, and other evidence remain attached to the work that produced them. Compute can run locally, over SSH, or on platforms such as Slurm, Kubernetes, Ray, Modal, and Hugging Face Jobs. Projects and experiment data remain local by default through a local dashboard and SQLite store. -
Demo Series - Small Business Backup By VeeamWatch this on-demand demo series and learn how to protect your Microsoft 365 data with clear, simple, actionable steps that are easy to implement for businesses of all sizes.
-
10
Shannon
Fully autonomous AI hacker to find actual exploits in your web apps
...The project blends source-aware analysis with automated web interaction so it can validate issues like injection flaws, authentication bypasses, and other exploitable paths in a way that resembles an actual attacker’s workflow. Instead of requiring you to manually reproduce findings, Shannon is designed to produce actionable evidence that a weakness can be weaponized, which helps teams prioritize what truly matters. It positions itself as a pre-attacker safety net, aiming to break your web app before someone else does and thereby reduce the gap between “potentially vulnerable” and “confirmed exploitable.” -
11
geo-sleuth
An agent skill that finds where a photo was taken
...Specialized Python scripts collect, score, and rank geographic candidates before the AI agent evaluates the strongest matches. Results include estimated coordinates, camera direction, confidence levels, and supporting visual evidence. Geo-Sleuth integrates with Claude Code, Codex, Cursor, Gemini CLI, OpenCode, and GitHub Copilot. -
12
Quint Code
Structured reasoning framework for Claude Code, Gemini, and Cursor
Quint Code is a structured reasoning and decision-support framework aimed at making AI-assisted software engineering and decision workflows more rigorous and auditable. It implements the First Principles Framework (FPF) to guide users and AI tools through hypothesis generation, logical verification, evidence gathering, and documented decision making, reducing reliance on ad hoc or “vibe” coding. Instead of accepting the first plausible answer generated by an AI assistant, Quint Code encourages generating multiple competing hypotheses, verifying them, and validating them against real evidence stored in a structured “knowledge base” within your project. ... -
13
reverse-skill
Reverse Engineering and Authorized Penetration Testing
...The package checks locally available tools and can bootstrap supporting scripts, MCP servers, and utilities when needed. Scope gates, case initialization, timelines, evidence chains, reports, and field journals encourage repeatable and auditable work. It supports clients such as Claude Code, Codex CLI, Cursor, and Cline across Windows, Linux, macOS, and Kali Linux. The repository also preserves reusable lessons so later investigations can avoid previously documented mistakes. -
14
VulnClaw
Based on AI Agent + MCP toolchain + penetration Skill orchestration
...Its newer architecture uses a goal-driven solving engine instead of a fixed-round loop, helping the agent stop when the goal is reached, the search space is exhausted, or a safety budget is met. It also includes evidence checks designed to reduce hallucinated conclusions by requiring real tool output before accepting key findings. VulnClaw is intended for authorized testing, CTFs, security education, and controlled red-team environments. -
15
REA: Reverse Engineer Anything
Reverse engineer anything with agents, from app behavior
REA (Reverse Engineer Anything) is an open-source, AI-powered reverse engineering toolkit that enables developers and AI agents to analyze software applications, native binaries, and runtime behavior. The platform uses the Model Context Protocol (MCP) to connect AI coding assistants with tools for inspecting compiled code, application structures, and software functionality. It supports reverse engineering of native executables, JavaScript and Electron applications, .NET assemblies, Android... -
16
CyberPPT
A Codex Skill for generating high-density, editable PowerPoints
CyberPPT is a Codex Skill for converting documents, research, and business data into dense, editable, consulting-style PowerPoint presentations. It extracts facts, numbers, claims, recommendations, conflicts, and caveats from formats such as PDF, DOCX, TXT, and XLSX. The workflow builds an MBB-style evidence table before developing storylines and converging on an SCR narrative. Users select from eight fixed visual systems, after which the skill plans each slide’s hierarchy, grid, charts, palette, and information density. Image-generation blueprints guide reconstruction while native text, shapes, tables, charts, and vectors preserve core editability. ... -
17
DeepSearcher
Open Source Deep Research Alternative to Reason and Search
...The overall workflow aims to minimize hallucinations by grounding outputs in retrieved material and then applying structured reasoning over that evidence before generating a final report. -
18
OpenRig
Multi-agent harness that runs Claude Code and Codex together
...The system runs on the user’s own infrastructure and keeps its orchestration visible through tmux. A command-line interface manages setup, health checks, queues, and workstation configuration. OpenRig also includes skills, evidence tracking, reusable packages, and support for persistent team workflows. It is aimed at developers who want several coding agents collaborating as a stable group rather than operating as isolated assistants. -
19
PaperSpine
Motivation-driven skill for learning from strong academic papers
...The project asks the agent to study target formats and strong examples before drafting or revising. It emphasizes the central motivation of a paper, helping writers connect claims, structure, evidence, citations, and revisions into a coherent argument. The suite includes specialized skills for research, citation, rewriting, LaTeX, auditing, translation, humanization, and updates. It is best suited for users who need format-aware, evidence-aware academic writing support rather than generic text generation. -
20
MetaScreener
AI-powered tool for efficient abstract and PDF screening
MetaScreener is an open-source AI-assisted tool designed to streamline the screening process in systematic literature reviews and academic research workflows. The system helps researchers analyze large collections of academic abstracts and research papers to determine which studies are relevant for inclusion in evidence synthesis projects. Instead of manually reviewing hundreds or thousands of documents, researchers can use MetaScreener to apply machine learning techniques that assist with classification and prioritization of candidate papers. The platform can analyze both abstracts and full PDF documents, enabling automated filtering based on research criteria defined by the user. ... -
21
anti-slop
Opinionated Oxlint rules for rejecting low-evidence TypeScript
Anti-slop is an opinionated Oxlint rule set for rejecting low-evidence TypeScript and JavaScript patterns. It is designed to be copied into a repository and customized rather than consumed as a fixed npm dependency. Its rules target practices such as chained type assertions, unsafe dictionary types, broad object parameters, ad hoc runtime type checks, module mocking, and contracts that return unknown. -
22
CampusShame
Crowdsource and document the reputational history of companies
...It compiles lists of companies that have had so-called “shameful” behaviors, such as rescinding offers, canceling intention letters, or terminating three-party agreements after acceptance, alongside a parallel list of companies without reported negative incidents. The project draws on publicly available anecdotes, snapshots, and forum posts from platforms such as NowCoder, Maimai, and Zhihu to provide evidence and context for each entry, preserving details that might otherwise be lost as threads get deleted or censored. It also includes commentary on recruitment pitfalls and encourages contributors to share their own experiences or corrections to keep the list up to date. -
23
Pi Web Access
Web search and content extraction extension for Pi coding agent
...GitHub repositories can be cloned locally so Pi can inspect real source files instead of relying on rendered pages. Video support includes transcripts, visual descriptions, timestamped frame extraction, and questions about on-screen content, while evidence tools help collect passages for manual claim verification. -
24
Loop Library
A library of practical AI-agent loops and an installable skill
...The repository contains both a public loop library website and the Loopy skill that agents can use to discover, audit, adapt, run, and prepare loops. It helps turn open-ended prompts into measurable workflows with evidence, stopping criteria, and human approval points. The skill can work with agent environments such as Codex, Cursor, and Claude Code through the Skills CLI. Overall, it is useful for teams that want safer, reusable, and more reliable agent workflows instead of one-off prompting. -
25
ResultTypes.jl
A Result type for Julia—it's like Nullables for Exceptions
ResultTypes provides a Result type that can hold either a value or an error. This allows us to return a value or an error in a type-stable manner without throwing an exception.
