1792 matches found
CVE-2026-57121
PraisonAI is a multi-agent teams system. From version 0.1.4 to before version 0.1.5 in praisonai-platform, a workspace member can permanently delete any resource — projects, agents, issues, labels, issue dependencies, and issue-label attachments — created by the workspace owner or other members...
CVE-2026-57113
PraisonAI is a multi-agent teams system. From version 2.6.0 to before version 4.6.59, PraisonAI's template loader accepts GitHub template URIs with refs, for example github:owner/repo/[email protected]. The resolver stores the user-controlled template path and ref verbatim, and the cache layer late...
CVE-2026-57117
PraisonAI is a multi-agent teams system. From version 4.6.10 to before version 4.6.59, LocalManagedAgent / SandboxedAgent compute bridging wraps readfile, listfiles, and writefile when a compute provider is attached. The bridge converts those file operations into shell command strings using raw...
CVE-2026-57146
The PraisonAI multi-agent system (versions 4.5.115 to 4.6.58 ) contains an incomplete authentication fix in the A2U Agent-to-User event stream server . When using the documented CLI entrypoint (praisonai serve a2u), the system fails to install API-key middleware or generate a default token. Inste...
CVE-2026-57144
The praisonai package (versions 4.5.110 through 4.6.58 ) is vulnerable to a Protection Mechanism Failure in its SandlockSandbox component. When the optional sandlock module is present but the host kernel lacks Landlock support, the execute() and run_command() functions fail open by silently falli...
CVE-2026-57142
The PraisonAI multi-agent system (versions >= 4.5.87 and
CVE-2026-57121
The praisonai-platform package (versions 0.1.4 to before 0.1.5) is vulnerable to an improper authorization issue in its Platform API. Specifically, six content DELETE endpoints for projects, agents, issues, labels, issue dependencies, and issue-label attachments fail to perform ownership or role ...
CVE-2026-57121 PraisonAI: Missing ownership check on DELETE endpoints allows members to delete others' content in Platform API
PraisonAI is a multi-agent teams system. From version 0.1.4 to before version 0.1.5 in praisonai-platform, a workspace member can permanently delete any resource — projects, agents, issues, labels, issue dependencies, and issue-label attachments — created by the workspace owner or other members...
CVE-2026-57116
The PraisonAI framework (package praisonai ) contains a critical authentication bypass in its AgentOS FastAPI deployment surface. Affected versions range from 4.2.1 through 4.6.57 , including the publicly reported patched version 4.5.128 . The root cause is that AgentOSConfig and AgentAppConfig d...
CVE-2026-57114
The PraisonAI multi-agent system (versions 4.5.126 through 4.6.57 ) is vulnerable to a Server-Side Request Forgery (SSRF) via DNS rebinding . The flaw exists in the Async Jobs API because the webhook_url is validated during request parsing and object construction, but this validation is not tied ...
CVE-2026-57113
PraisonAI versions 2.6.0 through 4.6.57 are vulnerable to a path traversal flaw in the template loader. The system accepts GitHub template URIs with refs but stores the user-controlled path and ref verbatim without normalization. Because the cache layer joins these values into a filesystem path w...
CVE-2026-57113 PraisonAI GitHub template cache path traversal allows outside-cache file write and directory deletion
PraisonAI is a multi-agent teams system. From version 2.6.0 to before version 4.6.59, PraisonAI's template loader accepts GitHub template URIs with refs, for example github:owner/repo/[email protected]. The resolver stores the user-controlled template path and ref verbatim, and the cache layer late...
CVE-2026-56840
The PraisonAI multi-agent system is vulnerable to Cross-site Scripting (XSS) in its HTTPApproval dashboard from versions 4.5.2 to before 4.6.59 . The root cause is the _build_html() function in src/praisonai/praisonai/bots/_http_approval.py, which uses raw f-string interpolation to render tool na...
CVE-2026-56838
PraisonAI versions 4.5.87 through 4.6.57 are vulnerable to a policy enforcement bypass in the recipe.run_stream() function. While the standard recipe.run() path correctly invokes _check_tool_policy() to block dangerous tools unless allow_dangerous_tools=True is specified, the streaming path calls...
CVE-2026-56837
The praisonai package (specifically the LinearBot component) is vulnerable to unauthenticated remote agent invocation in versions 4.5.50 and >= 4.6.56,
CVE-2026-56836
The PraisonAI multi-agent system (versions 4.5.112 through 4.6.58 ) contains a vulnerability where the deprecated Typer recipe serve command bypasses non-localhost authentication guards. Because the command fails to set the auth configuration, the create_app() function does not attach required AP...
CVE-2026-56835
The praisonai package (repository MervinPraison/PraisonAI) is vulnerable to an authorization bypass in its Slack bot integration. Specifically, the app_mention event handler (located in src/praisonai/praisonai/bots/slack.py) fails to enforce the allowed_users, allowed_channels, and unknown_user_p...
CVE-2026-56834
The PraisonAI multi-agent system (package praisonai) is vulnerable to arbitrary file read via its Dynamic Context Discovery artifact helper tools. Specifically, the artifact_head, artifact_tail, artifact_grep, and artifact_chunk functions in src/praisonai/praisonai/context/queue.py fail to enforc...
CVE-2026-56833
The PraisonAI multi-agent system (versions 3.8.1 through 4.6.58 ) is vulnerable to a path traversal issue in its Dynamic Context module. The vulnerability exists because the history_search, history_tail, history_get, terminal_tail, terminal_grep, and terminal_commands tools accept run_id and agen...
CVE-2026-56832
The PraisonAI multi-agent system (versions 4.5.2 through 4.6.58 ) contains an incorrect authorization vulnerability in its DiscordApproval backend. The praisonai.bots.DiscordApproval component improperly handles tool call approvals by polling the configured Discord channel for any non-bot message...