SuperAPI AI Directory
Partner with us中文AI newsFavorites

All AI tools

35,708 results · Page 896 of 992

Traccia

AI agents

Traccia is an AI agent control plane for production-grade AI systems, providing observability, evaluation, governance, and runtime policy enforcement. It connects to frameworks such as LangChain, CrewAI, OpenAI Agents SDK, AutoGen, and LlamaIndex through an OpenTelemetry-native SDK, and uses a unified dashboard to centrally manage all agents: view end-to-end traces, cost and token attribution for 2,500+ models, PII and sensitive data detection, prompt registry and experiment evaluation, and layered policy guardrail monitoring with hard blocking. Its distinctive features include the ability to enforce interception mid-execution of an agent (such as overspending, infinite loops, or PII leakage), cost accounting that is unaffected by sampling rates, and compliance evidence export mapped to the EU AI Act. It is suitable for AI teams that need unified observability, governance, and compliance auditing across frameworks.

Read more

tracefold

AI agents

TraceFold is a reversibility assurance tool for AI agents, with the core idea that AI agents should not make irreversible changes. It escrows the inverse operation before an effect lands, refuses the effect if it cannot be escrowed, and issues receipts that anyone can verify offline. TraceFold is not an observability tool; it sits before the effect occurs. The project installs a binary named gx, which is the CLI run in the demo, providing features such as gx undo and gx receipt verify. gx has been published on crates.io, version 0.1.2, under the Apache-2.0 license, and can also be built from source. TraceFold implements the Deterministic Approval Gate (Layer 3: Mechanical Law) and the Receipt & Escrow Substrate (Layer 8: Provenance and Receipts) within the broader Glovrex Digital World architecture. The project provides an offline verification webpage that verifies receipts and keys by running WebAssembly within a browser tab, with strictly zero network requests.

Read more

Traceloop

AI agents

Traceloop is a continuous feedback and observability platform for LLM applications, helping teams turn evaluation and monitoring into a closed loop of continuous improvement. It can be integrated with one line of code, tracks data such as prompts, responses, and latency in real time, and automatically runs quality checks such as faithfulness, relevance, and safety. It supports custom evaluators, automatically executes evaluations in CI or at runtime, is compatible with OpenTelemetry and the open-source OpenLLMetry SDK, and supports Python, TypeScript, Go, Ruby, as well as more than 20 model providers and mainstream frameworks. It offers cloud, on-premises, and isolated deployments, and complies with SOC 2 and HIPAA standards.

Read more

Tractable

AI productivity

Tractable is a technology company that applies artificial intelligence to vehicle and property damage assessment. Its official website describes its core positioning as "driving the future of auto assessment with intelligent automation," providing faster and more accurate damage assessments for repairers, recyclers, insurers, and others. Its AI is trained on applied research and millions of data-rich images, can perform pixel-level analysis of images, generate damage assessment results, and attach a certainty score to each valuation, factoring in elements such as visibility, image quality, and damage severity. Its product line covers AI damage detection and assessment within the claims process, AI imaging beyond damage detection, capturing more leads through AI, optimizing salvage bidding with AI-driven damage assessment and parts identification, and instant vehicle assessments for fleets and rentals, used to streamline pickups, returns, and fleet maintenance processes. According to the official website, the AI can process thousands of claims per day and identify recyclable parts during repairs; through an open API it can be integrated into existing systems; it also continuously learns and improves as it is deployed. The official website also mentions LumaScanner, calling it the latest innovation for improving fixed operations, but does not provide specific feature details. Tractable has a global network of industry customers, and the official website showcases collaboration cases such as Kirmac Collision, Foyer, and Covéa, as well as the latest updates.

Read more

TradingAgents-CN

AI finance

TradingAgents-CN is a Chinese-language financial trading research framework based on multi-agent large language models (LLMs), oriented toward the Chinese A-share market and providing stock research assistance functions. The system collaborates through multiple AI agents playing different roles, including market-wide, sector, market, fundamental, and news analysts, as well as optimistic and prudent researchers, completing the research process through debate-style analysis. Main features include single-stock research (supporting 5-level deep multi-agent debate-style analysis), general research (asking questions in natural language to research any market topic), stock screening (natural language plus conditional screening, covering 6,000+ stocks), stock selection assistant (matching candidate stock pools according to research needs), intelligent assistant (web search, tool invocation, and factual memory), AI memory, Skill Center, and Learning Center. The project adopts a hybrid licensing model: the core engine is open source under Apache-2.0, while the application layer is proprietary source-available. Deployment requires self-installing MongoDB 7.0+ and Redis 6+; the backend automatically initializes system configuration and creates an administrator account on first startup. This tool is for research assistance purposes; all AI-generated content is for learning reference only and does not constitute investment advice.

Read more