Building a RAG agent usually means stitching together a retriever, a workflow engine and a model layer by hand. FastGPT from is an open-source agent platform with a visual workflow builder, knowledge base hybrid retrieval with reranking, and full call-chain logs for debugging. It has a free...
November 2, 2026 at 3:14 PM
Building a RAG agent usually means stitching together a retriever, a workflow engine and a model layer by hand. FastGPT from is an open-source agent platform with a visual workflow builder, knowledge base hybrid retrieval with reranking, and full call-chain logs for debugging. It has a free...
Evidence, not compliance — no word on whether it maps to what CAC reviewers request. Still, self-hosting an open-source stack means you can verify what gets logged. Hard to see FastGPT's format replacing the hand-rolled exports until one inspection proves it holds.
AI Content Moderation Compliance Tightens as FastGPT Launches Standardised Moderation Audit Trail
FastGPT has introduced a standardised content moderation and audit-trail scheme for enterprise AI deployments, aiming to clarify responsibility, configure sampling ratios and retain full-chain records as generative AI compliance requirements tighten.
aichina.news
October 13, 2026 at 12:27 AM
Evidence, not compliance — no word on whether it maps to what CAC reviewers request. Still, self-hosting an open-source stack means you can verify what gets logged. Hard to see FastGPT's format replacing the hand-rolled exports until one inspection proves it holds.
🚨 EUVD-2026-84559
📊 6.3/10
🏢 labring
📝 FastGPT is an open-source LLM platform for building AI applications on a knowledge base. Prior to 4.15.2, the safe Axios request interceptor in packages/s...
🔗 https://euvd.enisa.europa.eu/vulnerability/EUVD-2026-84559
#cybersecurity #infosec #cve #euvd
📊 6.3/10
🏢 labring
📝 FastGPT is an open-source LLM platform for building AI applications on a knowledge base. Prior to 4.15.2, the safe Axios request interceptor in packages/s...
🔗 https://euvd.enisa.europa.eu/vulnerability/EUVD-2026-84559
#cybersecurity #infosec #cve #euvd
September 22, 2026 at 6:02 PM
🚨 EUVD-2026-84559
📊 6.3/10
🏢 labring
📝 FastGPT is an open-source LLM platform for building AI applications on a knowledge base. Prior to 4.15.2, the safe Axios request interceptor in packages/s...
🔗 https://euvd.enisa.europa.eu/vulnerability/EUVD-2026-84559
#cybersecurity #infosec #cve #euvd
📊 6.3/10
🏢 labring
📝 FastGPT is an open-source LLM platform for building AI applications on a knowledge base. Prior to 4.15.2, the safe Axios request interceptor in packages/s...
🔗 https://euvd.enisa.europa.eu/vulnerability/EUVD-2026-84559
#cybersecurity #infosec #cve #euvd
FastGPT's Latest Updates: Enhanced Runtime and Deployment Guidance for AI Agents#China#Hangzhou#AI_Agents#FastGPT#Agent_Sandbox
FastGPT's Latest Updates: Enhanced Runtime and Deployment Guidance for AI Agents
FastGPT has released new guidance on Agent Sandbox runtime and deployment, detailing critical changes for user configurations and operational updates.
third-news.com
September 14, 2026 at 11:15 AM
FastGPT's Latest Updates: Enhanced Runtime and Deployment Guidance for AI Agents#China#Hangzhou#AI_Agents#FastGPT#Agent_Sandbox
📢 New #updates · 11 Sept #10
· MedusaJs v2.21.0 — Security hardening for Store API routes
· fastgpt v4.17.0 — Automated upgrades with lease-based concurrency control
· MedusaJs v2.21.0 — Security hardening for Store API routes
· fastgpt v4.17.0 — Automated upgrades with lease-based concurrency control
September 11, 2026 at 11:38 AM
📢 New #updates · 11 Sept #10
· MedusaJs v2.21.0 — Security hardening for Store API routes
· fastgpt v4.17.0 — Automated upgrades with lease-based concurrency control
· MedusaJs v2.21.0 — Security hardening for Store API routes
· fastgpt v4.17.0 — Automated upgrades with lease-based concurrency control
🔒 Read the Full Story
FastGPT Documents Data Retention Periods and Deletion Semantics
FastGPT, an open-source AI application platform for organizations, has consolidated in its public documentation how four classes of data are retained and removed: conversation records, knowledge base files, model-call traces…
techintelpro.com
September 8, 2026 at 2:33 PM
🔒 Read the Full Story
New Enhancements in Workflow and Deployment for FastGPT AI Orchestration Platform#China#Hangzhou#Open_Source#AI_Orchestration#FastGPT
New Enhancements in Workflow and Deployment for FastGPT AI Orchestration Platform
FastGPT has introduced significant updates, including v4.16.0, focusing on enhancing workflow design and deployment processes for its open-source AI platform.
third-news.com
September 8, 2026 at 1:08 AM
New Enhancements in Workflow and Deployment for FastGPT AI Orchestration Platform#China#Hangzhou#Open_Source#AI_Orchestration#FastGPT
FastGPT Clarifies How AI Data Management Handles Deletion and Retention Policies#China#Hangzhou#AI_Platform#Data_Management#FastGPT
FastGPT Clarifies How AI Data Management Handles Deletion and Retention Policies
FastGPT has released a comprehensive guide on data retention and deletion policies, clarifying how user data is handled within its AI platform.
third-news.com
September 4, 2026 at 6:05 AM
FastGPT Clarifies How AI Data Management Handles Deletion and Retention Policies#China#Hangzhou#AI_Platform#Data_Management#FastGPT
FastGPT 4.15.1 WeChat Channel Authorization Bypass https://packetstorm.news/files/230085 #exploit
August 31, 2026 at 8:28 PM
FastGPT 4.15.1 WeChat Channel Authorization Bypass https://packetstorm.news/files/230085 #exploit
🚨 CVE-2026-68929 — CVSS 9.3 CRITICAL
FastGPT is an open-source LLM platform for building AI applications on a knowledge base. In versions prior ...
🔎 https://stemshop.top/cve/CVE-2026-68929
#CVE #CyberSecurity #InfoSec
FastGPT is an open-source LLM platform for building AI applications on a knowledge base. In versions prior ...
🔎 https://stemshop.top/cve/CVE-2026-68929
#CVE #CyberSecurity #InfoSec
August 28, 2026 at 2:07 AM
🚨 CVE-2026-68929 — CVSS 9.3 CRITICAL
FastGPT is an open-source LLM platform for building AI applications on a knowledge base. In versions prior ...
🔎 https://stemshop.top/cve/CVE-2026-68929
#CVE #CyberSecurity #InfoSec
FastGPT is an open-source LLM platform for building AI applications on a knowledge base. In versions prior ...
🔎 https://stemshop.top/cve/CVE-2026-68929
#CVE #CyberSecurity #InfoSec
FastGPT Releases Version 4.16.0: Enhanced Sandboxes and Metadata Features for Enterprise Use#China#Hangzhou#Enterprise_Solutions#Open-Source_AI#FastGPT
FastGPT Releases Version 4.16.0: Enhanced Sandboxes and Metadata Features for Enterprise Use
FastGPT's latest release v4.16.0 introduces notable enhancements including user-level code sandboxes, customizable metadata, and improved audit log retention strategies. Explore these updates.
third-news.com
August 20, 2026 at 5:55 AM
FastGPT Releases Version 4.16.0: Enhanced Sandboxes and Metadata Features for Enterprise Use#China#Hangzhou#Enterprise_Solutions#Open-Source_AI#FastGPT
I mean, you can use Kagi without any of the AI stuff. The Quick Answer (FastGPT) doesn't even show up unless you press the button or end the query in a "?". They're pretty good at not slapping it in people's faces.
January 16, 2026 at 7:54 PM
I mean, you can use Kagi without any of the AI stuff. The Quick Answer (FastGPT) doesn't even show up unless you press the button or end the query in a "?". They're pretty good at not slapping it in people's faces.
Kagi eased me into the good side of AI. I was very anti-AI even when I began using Kagi, point blank refusing to even use FastGPT, but have slowly come to appreciate the way you go about it, and now I can use it appropriately. A company being pro-AI but anti-slop is perfect 🔥
January 15, 2026 at 4:53 PM
Kagi eased me into the good side of AI. I was very anti-AI even when I began using Kagi, point blank refusing to even use FastGPT, but have slowly come to appreciate the way you go about it, and now I can use it appropriately. A company being pro-AI but anti-slop is perfect 🔥
Kagi, my beloved, listing the actual sources used during the AI answer (FastGPT), which only even triggers when the query ends in "?"
This is how they should *all* work
This is how they should *all* work
December 18, 2025 at 12:06 AM
Kagi, my beloved, listing the actual sources used during the AI answer (FastGPT), which only even triggers when the query ends in "?"
This is how they should *all* work
This is how they should *all* work
Jeg bruger kagi.com som søgemaskine. Den koster penge, men til gengæld er jeg ikke produktet. Den har også en sprogmodel, FastGPT, som kun giver svar med kildehenvisning.
Kagi Search - A Premium Search Engine
Better search results with no ads. Welcome to Kagi (pronounced kah-gee), a paid search engine that gives power back to the user.
kagi.com
February 2, 2025 at 1:49 PM
Jeg bruger kagi.com som søgemaskine. Den koster penge, men til gengæld er jeg ikke produktet. Den har også en sprogmodel, FastGPT, som kun giver svar med kildehenvisning.
i assume the reason the last three all say 24mph is because they're using the same model and training data, just with their own tweaks. it's interesting that fastgpt hallucinated an answer from the movie, which doesn't provide one. bard seems to have the best grasp on the original context.
January 3, 2024 at 4:48 PM
i assume the reason the last three all say 24mph is because they're using the same model and training data, just with their own tweaks. it's interesting that fastgpt hallucinated an answer from the movie, which doesn't provide one. bard seems to have the best grasp on the original context.
google bard vs. microsoft copilot vs. openai chatgpt vs. kagi fastgpt
January 3, 2024 at 4:40 PM
google bard vs. microsoft copilot vs. openai chatgpt vs. kagi fastgpt
CVE-2026-50562 - fastgpt
FastGPT's preview workflows can accidentally push malicious Docker images to GHCR, potentially exposing sensitive data. This is a concern because attackers could exploit this to gain…
Too many irrelevant or confusing CVEs? Use stackflag.com
#fastgpt #labring #CVE #infosec
FastGPT's preview workflows can accidentally push malicious Docker images to GHCR, potentially exposing sensitive data. This is a concern because attackers could exploit this to gain…
Too many irrelevant or confusing CVEs? Use stackflag.com
#fastgpt #labring #CVE #infosec
CVE-2026-50562: FastGPT: Malicious Docker Images Can Be Pushed to GHCR
FastGPT's preview workflows can accidentally push malicious Docker images to GHCR, potentially exposing sensitive data.
stackflag.com
July 15, 2026 at 6:42 PM
CVE-2026-50562 - fastgpt
FastGPT's preview workflows can accidentally push malicious Docker images to GHCR, potentially exposing sensitive data. This is a concern because attackers could exploit this to gain…
Too many irrelevant or confusing CVEs? Use stackflag.com
#fastgpt #labring #CVE #infosec
FastGPT's preview workflows can accidentally push malicious Docker images to GHCR, potentially exposing sensitive data. This is a concern because attackers could exploit this to gain…
Too many irrelevant or confusing CVEs? Use stackflag.com
#fastgpt #labring #CVE #infosec
I should point out that Kagi search is what I am referring to above. Their Fastgpt is the search product. But you can also upgrade to use the Kagi Assistant that allows you to choose many of the AI providers and privides privacy protections. help.kagi.com/kagi/ai/assi...
help.kagi.com
November 29, 2024 at 3:04 AM
I should point out that Kagi search is what I am referring to above. Their Fastgpt is the search product. But you can also upgrade to use the Kagi Assistant that allows you to choose many of the AI providers and privides privacy protections. help.kagi.com/kagi/ai/assi...
Well I’m stuck with it until next year as I paid for an annual sub. If search goes bad then, I’ll go elsewhere. But for now it suits my needs for a no nonsense search. I’ve been aware of their FastGPT and summarizer tools but I’m not using them. The good outweighs the bad for now.
May 14, 2024 at 6:05 PM
Well I’m stuck with it until next year as I paid for an annual sub. If search goes bad then, I’ll go elsewhere. But for now it suits my needs for a no nonsense search. I’ve been aware of their FastGPT and summarizer tools but I’m not using them. The good outweighs the bad for now.
🟠 CVE-2026-44285 - High (7.7)
FastGPT is an AI Agent building platform. Prior to 4.15.0-beta1, a Server-Side Request Forgery (S...
https://www.thehackerwire.com/vulnerability/CVE-2026-44285/
#infosec #cybersecurity #CVE #vulnerability #security #patchstack
FastGPT is an AI Agent building platform. Prior to 4.15.0-beta1, a Server-Side Request Forgery (S...
https://www.thehackerwire.com/vulnerability/CVE-2026-44285/
#infosec #cybersecurity #CVE #vulnerability #security #patchstack
May 29, 2026 at 9:01 PM
🟠 CVE-2026-44285 - High (7.7)
FastGPT is an AI Agent building platform. Prior to 4.15.0-beta1, a Server-Side Request Forgery (S...
https://www.thehackerwire.com/vulnerability/CVE-2026-44285/
#infosec #cybersecurity #CVE #vulnerability #security #patchstack
FastGPT is an AI Agent building platform. Prior to 4.15.0-beta1, a Server-Side Request Forgery (S...
https://www.thehackerwire.com/vulnerability/CVE-2026-44285/
#infosec #cybersecurity #CVE #vulnerability #security #patchstack
🔴 CVE-2026-42302 - Critical (9.8)
FastGPT is an AI Agent building platform. From version 4.14.10 to before version 4.14.13, the age...
https://www.thehackerwire.com/vulnerability/CVE-2026-42302/
#infosec #cybersecurity #CVE #vulnerability #security #patchstack
FastGPT is an AI Agent building platform. From version 4.14.10 to before version 4.14.13, the age...
https://www.thehackerwire.com/vulnerability/CVE-2026-42302/
#infosec #cybersecurity #CVE #vulnerability #security #patchstack
May 8, 2026 at 11:25 PM
🔴 CVE-2026-42302 - Critical (9.8)
FastGPT is an AI Agent building platform. From version 4.14.10 to before version 4.14.13, the age...
https://www.thehackerwire.com/vulnerability/CVE-2026-42302/
#infosec #cybersecurity #CVE #vulnerability #security #patchstack
FastGPT is an AI Agent building platform. From version 4.14.10 to before version 4.14.13, the age...
https://www.thehackerwire.com/vulnerability/CVE-2026-42302/
#infosec #cybersecurity #CVE #vulnerability #security #patchstack
🟠 CVE-2026-42345 - High (7.7)
FastGPT is an AI Agent building platform. In versions 4.14.11 and prior, FastGPT's isInternalAddr...
https://www.thehackerwire.com/vulnerability/CVE-2026-42345/
#infosec #cybersecurity #CVE #vulnerability #security #patchstack
FastGPT is an AI Agent building platform. In versions 4.14.11 and prior, FastGPT's isInternalAddr...
https://www.thehackerwire.com/vulnerability/CVE-2026-42345/
#infosec #cybersecurity #CVE #vulnerability #security #patchstack
May 8, 2026 at 11:24 PM
🟠 CVE-2026-42345 - High (7.7)
FastGPT is an AI Agent building platform. In versions 4.14.11 and prior, FastGPT's isInternalAddr...
https://www.thehackerwire.com/vulnerability/CVE-2026-42345/
#infosec #cybersecurity #CVE #vulnerability #security #patchstack
FastGPT is an AI Agent building platform. In versions 4.14.11 and prior, FastGPT's isInternalAddr...
https://www.thehackerwire.com/vulnerability/CVE-2026-42345/
#infosec #cybersecurity #CVE #vulnerability #security #patchstack