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/service/common/api/axios.ts validates a hostname with isInternalAddress() before a later HTTP connection performs an independent DNS lookup, creating a DNS rebinding window, allowing an attacker-controlled hostname to resolve publicly during the check and to a loopback, private, link-local, or metadata address during connection. An authenticated attacker who can supply a URL to a safe-Axios-backed HTTP tool, workflow HTTP node, external file fetch, or other server-side integration can use this time-of-check/time-of-use gap to reach services that direct private-address inputs would block. The same independent re-resolution occurs after manual redirect hops because each redirect target is checked before a separate connection lookup. This issue is fixed in version 4.15.2.
FastGPT versions prior to 4.15.2 have a missing authorization vulnerability in the WeChat (iLink) share-channel endpoints. These endpoints rely solely on a public shareId for authorization, which is exposed in shared URLs and embeds, allowing unauthenticated attackers to hijack or disrupt victim teams' WeChat bots. Attackers can log out the victim's bot or bind the victim's app to their own bot, exposing private responses and consuming victim resources. This issue is fixed in version 4.15.2.
FastGPT is a knowledge-based AI application platform. From 4.14.17 until 4.15.0-beta5, an authenticated FastGPT user can save a workflow node that points to another user's private HTTP toolset by using a crafted saved tool id such as http-<victim_toolset_app_id>/<tool_name>. The normal toolset routes deny access, but the workflow save and runtime path did not apply the same authorization check to the referenced toolset, allowing /api/v2/chat/completions to resolve the saved reference and execute the victim-owned HTTP tool. This issue is fixed in version 4.15.0-beta5.
FastGPT is a knowledge-based AI application platform. At commit 22ebfacbb43311e9b73294040ae0eb87390c6bba and earlier, artifacts built from untrusted pull request code in .github/workflows/preview-docs-build.yml and .github/workflows/preview-fastgpt-build.yml can be downloaded by privileged workflow_run jobs in .github/workflows/preview-docs-push.yml and .github/workflows/preview-fastgpt-push.yml, allowing attacker-controlled Docker images from the document/ tree or FastGPT build context to be pushed to GHCR and, for documentation previews, deployed with secrets.KUBE_CONFIG_CN.
FastGPT is a knowledge-based AI application platform. Prior to 4.15.0-beta5, FastGPT's shared SSRF guard validates only the initial request URL before handing the request to axios, and axios follows redirects by default. An authenticated workflow user can configure an HTTP request node to call an attacker-controlled public URL that redirects to cloud metadata, loopback, or internal services that the guard would block on direct request, and the HTTP node returns the response body to the workflow caller. This issue is fixed in version 4.15.0-beta5.
FastGPT is a knowledge-based AI application platform. From 4.14.17 until 4.15.0-beta5, the POST /api/core/chat/record/getCollectionQuote endpoint authenticates the caller's chat and collection context, but the initialId center-node lookup is not bound to that authorized context. A low-privileged tenant user can call the endpoint with valid attacker-owned appId, chatId, chatItemDataId, and collectionId values while supplying another tenant's dataset data id as initialId, causing the response to include foreign dataset quote or full-text content. This issue is fixed in version 4.15.0-beta5.
FastGPT is a knowledge-based AI application platform. In 4.15.0-beta4, FastGPT plugin invoke reverse-call endpoints under /api/invoke/* authenticate only by verifying a JWT signed with INVOKE_TOKEN_SECRET, which defaults to the constant string token and was not set in official deployment templates. An unauthenticated attacker can self-sign an HS256 JWT and reach /api/invoke/userInfo to disclose cross-tenant user PII by attacker-supplied tmbId values, or /api/invoke/fileUpload to write attacker-controlled content into chat files. This issue is fixed in version 4.15.0-beta5.
FastGPT is a knowledge-based AI application platform. Prior to 4.15.0, GET /api/core/ai/record/getRecord authenticates the caller but loads LLM request and response traces only by requestId without team scoping, allowing any authenticated user to read another team's prompts, retrieved RAG chunks, and completions if the requestId is known. This issue is fixed in version 4.15.0.
FastGPT is a knowledge-based AI application platform. Prior to 4.15.0-beta4, the HTTP-tool OpenAPI schema importer validates only the top-level URL before passing it to SwaggerParser.bundle, whose remote reference resolver fetches $ref URLs without FastGPT's internal-address guard and returns fetched content inline, allowing an authenticated team member to read internal services or cloud metadata. This issue is fixed in version 4.15.0-beta4.
FastGPT is an open source AI knowledge base platform. Prior to v4.15.0-beta5, two FastGPT file handlers authorize an unrelated resource and then sign or read an S3 object using a key taken directly from the request, without checking that the key belongs to the caller's team. Because S3 object keys are global within the bucket and carry the tenant id only as a path segment, an attacker can supply another team's key and obtain its file contents through the chat-file presign endpoint or dataset preview endpoint. This issue is fixed in version v4.15.0-beta5.
FastGPT is an open source AI knowledge base platform. From 4.14.17 to before 4.15.0-beta4, FastGPT allows an authenticated tenant user to call POST /api/core/dataset/collection/create/reTrainingCollection in a way that persists a server-owned datasetId value from another tenant. This creates mixed dataset objects and downstream dataset, collection, and training endpoints then make authorization decisions from inconsistent ownership anchors, allowing cross-tenant read, update, and delete access when mixed object ids are known. This issue is fixed in version 4.15.0-beta4.
FastGPT is an AI Agent building platform. Prior to 4.15.0-beta1, the JavaScript sandbox worker at projects/code-sandbox/src/pool/worker.ts:356 blocks dynamic import() with the regex /\bimport\s*\(/.test(code). JavaScript syntax accepts a block comment between import and (; the regex matches only ASCII whitespace, and the bytes /, *, *, / are not in the \s character class. The payload import/**/("child_process") parses as a syntactically valid dynamic import that the regex does not detect. Because import() is not wrapped by the safeRequire Proxy (which only proxies require), the attacker loads child_process and calls execSync - arbitrary command execution as uid=100(sandbox) inside the sandbox container. This vulnerability is fixed in 4.15.0-beta1.
FastGPT is an AI Agent building platform. Prior to 4.15.0-beta1, a Server-Side Request Forgery (SSRF) vulnerability allows an authenticated attacker to bypass the global isInternalAddress network protection and make arbitrary HTTP GET requests to internal network services. This is achieved by exploiting an incomplete fix in the dataset preview endpoint /api/core/dataset/file/getPreviewChunks when utilizing the externalFile data import type. This vulnerability is fixed in 4.15.0-beta1.
FastGPT is an AI Agent building platform. Prior to version 4.14.17, an unauthenticated Server-Side Request Forgery (SSRF) vulnerability allows attackers (or authenticated users with App editing privileges) to send arbitrary HTTP requests to internal/private network addresses. The fetchData function in the lafModule workflow node uses axios to fetch user-controlled URLs without validating them against the application's internal network blocklist guard (isInternalAddress), bypassing SSRF protections. This issue has been patched in version 4.14.17.
FastGPT is an AI Agent building platform. Prior to version 4.14.17, FastGPT had an inconsistent SSRF protection gap in MCP tool URL handling. The direct MCP preview/run endpoints already rejected internal/private network URLs, but the MCP tool create/update endpoints could still save an internal MCP server URL. That stored URL could later be used by workflow execution without revalidating the destination. An authenticated user with permission to create or manage MCP toolsets could store an internal endpoint such as http://localhost:3000/mcp and later cause the FastGPT backend workflow runner to connect to that internal destination. This issue has been patched in version 4.14.17.
FastGPT is an AI Agent building platform. In versions 4.14.11 and prior, FastGPT's isInternalAddress() function in packages/service/common/system/utils.ts blocks cloud metadata endpoints using a fullUrl.startsWith() check against a hardcoded list. This check can be bypassed using at least 7 different URL encoding techniques, all of which resolve to the same cloud metadata service but do not match the blocklist patterns. Additionally, the broader private IP check (isInternalIPv4/isInternalIPv6) is disabled by default because CHECK_INTERNAL_IP defaults to false (not 'true'), so these bypasses reach the metadata endpoint without any further validation. At time of publication, there are no publicly available patches.
FastGPT is an AI Agent building platform. In versions 4.14.11 and prior, FastGPT's isInternalAddress() function in packages/service/common/system/utils.ts is vulnerable to DNS rebinding (TOCTOU — Time-of-Check to Time-of-Use). The function resolves the hostname via dns.resolve4()/dns.resolve6() and checks resolved IPs against private ranges, but the actual HTTP request happens in a separate call with a new DNS resolution, allowing the DNS record to change between validation and fetch. At time of publication, there are no publicly available patches.
FastGPT is an AI Agent building platform. In versions 4.14.13 and prior, the code-sandbox component suffers from insufficient resource isolation and uncontrolled resource consumption. The service relies solely on an application-level soft limit (a 500ms polling interval) for memory management and lacks strict OS-level constraints such as cgroups or kernel-level namespaces. This architectural weakness allows attackers to easily bypass memory checks via time-window attacks, or exhaust the entire JavaScript worker pool via concurrent CPU-intensive requests, resulting in a complete Denial of Service (DoS) for legitimate users. At time of publication, there are no publicly available patches.
FastGPT is an AI Agent building platform. From version 4.14.10 to before version 4.14.13, the agent-sandbox component of FastGPT is vulnerable to unauthenticated Remote Code Execution (RCE). The startup script entrypoint.sh initializes code-server with the --auth none flag and binds the service to all network interfaces (0.0.0.0:8080). This configuration allows any user with network access to the port to bypass authentication and gain full control over the sandbox environment. This issue has been patched in version 4.14.13.
FastGPT is an AI Agent building platform. In versions prior to 4.14.9.5, the password change endpoint is vulnerable to NoSQL injection. An authenticated attacker can bypass the "old password" verification by injecting MongoDB query operators. This allows an attacker who has gained a low-privileged session to change the password of their account (or others if combined with ID manipulation) without knowing the current one, leading to full account takeover and persistence. This issue has been fixed in version 4.14.9.5.