CISA list

● PyPI package

praisonaiagents

Praisonaiagents has 50 security advisories. The newest release any of them names as the fix is 4.5.128.

As of , praisonaiagents has 50 security advisories. The newest release any of them names as the fix is 4.5.128.

Your version, and the version to upgrade to

Find the row your installed version falls in. The second column is the first release after it that none of the advisories lists as affected; a release outside every row is one no advisory lists. That is what the advisories say, not a promise that a release is free of flaws.
If you runUpgrade toAdvisories
Before 4.5.1284.5.12850 advisories

Security advisories, newest first

Security advisories, newest first
AdvisoryWhat it isAffectsFixed inPublished
CVE-2026-61443GHSA-c44f-37qr-gw3f, PYSEC-2026-4270PraisonAI: SkillTools Executes Scripts Without Path Containment ValidationHigh severityBefore 1.6.781.6.78
CVE-2026-61437GHSA-4gfv-wg42-7jw5, PYSEC-2026-4267PraisonAI: Unsafe Dynamic Module Loading Leads to Arbitrary Code Execution via tools.py in AgentFlowHigh severityBefore 1.6.781.6.78
CVE-2026-61432GHSA-4xxv-6wmf-xf45, PYSEC-2026-4268PraisonAI: FastContext path resolution permits absolute and traversal reads outside the workspaceModerate severityBefore 1.6.781.6.78
CVE-2026-61446GHSA-m6wp-h223-4c8g, PYSEC-2026-4271PraisonAI: Plugin Auto-Discovery Executes Arbitrary Python Files Without VerificationHigh severityBefore 1.6.781.6.78
CVE-2026-61447GHSA-2xv2-w8cq-5gxw, PYSEC-2026-4266PraisonAI: CodeAgent Executes LLM-Generated Code Without Sandboxing and Leaks All Environment SecretsCritical severityBefore 1.6.781.6.78
CVE-2026-60087GHSA-29r9-67vg-qj56, PYSEC-2026-4265PraisonAI: Human-in-the-loop tool approval is cached by tool name and silently reused for all subsequent calls with arbitrary argumentsModerate severityBefore 1.6.781.6.78
CVE-2026-61430GHSA-qg25-6gc4-48mg, PYSEC-2026-4272PraisonAI: DNS rebinding bypass in `web_crawl` SSRF protection allows internal response disclosureHigh severityBefore 1.6.781.6.78
CVE-2026-60089GHSA-qjw5-xwrp-xwpq, PYSEC-2026-4273PraisonAI: Project config can auto-save agent output outside the project rootModerate severityBefore 1.6.781.6.78
CVE-2026-61429GHSA-6g59-gm2v-qhvq, PYSEC-2026-4269PraisonAI: Crawl4AI/Chromium backend is also affected by the `web_crawl` SSRF validation bypassHigh severityBefore 1.6.781.6.78
CVE-2026-55527GHSA-gxmw-5f7x-6g22, PYSEC-2026-3902praisonaiagents vulnerable to arbitrary file write via unsanitized `user_id` in `FileMemory.__init__()`, path traversal to any writable locationHigh severityBefore 1.6.581.6.58
CVE-2026-55530GHSA-cfxv-8fw8-rwpv, PYSEC-2026-3901praisonaiagents: ast_grep_rewrite rewrites arbitrary files without the @require_approval gate enforced on every sibling mutation toolModerate severityBefore 1.6.581.6.58
CVE-2026-55524GHSA-vg6p-v9vm-6fgj, PYSEC-2026-3904praisonaiagents vulnerable to SSRF in web_crawl tool via redirect-following and DNS rebinding (validate-then-fetch gap)High severityBefore 1.6.581.6.58
CVE-2026-55526GHSA-x44h-65qv-cw74, PYSEC-2026-3905praisonaiagents has an SSRF protection bypass in `spider_tools._host_is_blocked()` via DNS-resolved hostnames (`127.0.0.1.nip.io`)High severityBefore 1.6.581.6.58
CVE-2026-55528GHSA-7g3p-92qq-8wvh, PYSEC-2026-3899praisonaiagents: AgentServer declares auth_token but never enforces it on any routeHigh severityBefore 1.6.581.6.58
CVE-2026-55523GHSA-8hjw-25cg-g52h, PYSEC-2026-3900praisonaiagents has a `web_crawl` SSRF protection bypass via unchecked redirect targetsHigh severityFrom 1.5.128 before 1.6.581.6.58
CVE-2026-55522GHSA-hxmv-c4g6-5fqc, PYSEC-2026-3891, PYSEC-2026-3903PraisonAI workflow include bypasses tools.py autoload opt-in and executes included recipe codeHigh severityFrom 0.12.12 before 1.6.581.6.58
CVE-2026-55525GHSA-5r34-2g38-6569, PYSEC-2026-3898praisonaiagents web_crawl vulnerable to SSRF via redirect-followingHigh severityBefore 1.6.581.6.58
CVE-2026-57143GHSA-4pcv-mg8v-vrgf, PYSEC-2026-3530PraisonAI: Server-Side Request Forgery (SSRF) in SearxNG / search_web tools via attacker-controlled searxng_url parameterHigh severityBefore 1.6.611.6.61
CVE-2026-57130GHSA-c969-5x3p-vq3v, PYSEC-2026-3532PraisonAI: IMAP Command Injection via Unsanitized Email Search ParametersHigh severityBefore 1.6.591.6.59
CVE-2026-57118GHSA-x8cv-xmq7-p8xp, PYSEC-2026-3537PraisonAI AgentTeam.launch exposes unauthenticated remote agent listing and invocation endpointsCritical severityBefore 1.6.591.6.59
CVE-2026-57129GHSA-2rcg-mm5h-xchx, PYSEC-2026-3527PraisonAI: Arbitrary File Read via `@file:` Mention Path TraversalHigh severityBefore 1.6.591.6.59
CVE-2026-57126GHSA-vxgj-xg5c-p4h7, PYSEC-2026-3535praisonaiagents: SSRF guard validates literal IPs only and never resolves DNSHigh severityBefore 1.6.591.6.59
CVE-2026-57125GHSA-4869-x4pr-q22x, PYSEC-2026-3501, PYSEC-2026-3529PraisonAI: Unauthenticated RCE via Jobs API + Approval BypassCritical severityBefore 1.6.591.6.59
CVE-2026-57123GHSA-x227-pf99-vffg, PYSEC-2026-3536PraisonAI: MCP SSE transport binds 0.0.0.0 with no authentication and no Origin validation; bundled SecurityConfig is never wired inCritical severityBefore 1.6.591.6.59
CVE-2026-57120GHSA-pv2j-rghr-v5r9, PYSEC-2026-3533PraisonAI: execute_code sandbox bypass: str.format C-level attribute access reads every blocklisted dunderModerate severityBefore 1.6.591.6.59
CVE-2026-57115GHSA-6h9p-93hq-q7h6, PYSEC-2026-3531PraisonAI: SpiderTools redirect-target SSRF protection bypassModerate severityBefore 1.6.591.6.59
CVE-2026-57112GHSA-vmf9-xx9w-86wx, PYSEC-2026-3521, PYSEC-2026-3534PraisonAI ToolsMCPServer legacy SSE transport accepts attacker Host/Origin and exposes registered toolsHigh severityFrom 0.6.0 before 1.6.591.6.59
CVE-2026-57128GHSA-35w5-pcw4-jx94, PYSEC-2026-3528PraisonAI: Unauthenticated Event Injection via SSE `/publish` EndpointModerate severityBefore 1.6.591.6.59
CVE-2026-47392GHSA-4mr5-g6f9-cfrh, PYSEC-2026-463, PYSEC-2026-483PraisonAI vulnerable to sandbox escape via `print.__self__` builtins module leak in `execute_code` (subprocess mode)Critical severityBefore 1.6.401.6.40
CVE-2026-47395GHSA-5cxw-77wg-jrf3, PYSEC-2026-2903, PYSEC-2026-2942PraisonAI CLI automatically resolves @url mentions in prompt text and can read loopback URLs into model contextModerate severityBefore 1.6.401.6.40
CVE-2026-47390GHSA-5c6w-wwfq-7qqm, PYSEC-2026-2902, PYSEC-2026-2941PraisonAI spider_tools SSRF protection bypass via alternate loopback host encodingsModerate severityBefore 1.6.401.6.40
CVE-2026-44339GHSA-gmjg-hv98-qggq, PYSEC-2026-2915, PYSEC-2026-2948PraisonAI has unsafe tool resolution in `ToolExecutionMixin.execute_tool`: undeclared `__main__` callables executeHigh severityBefore 1.6.371.6.37
CVE-2026-44335GHSA-q9pw-vmhh-384g, PYSEC-2026-2950PraisonAI has an SSRF bypassHigh severityBefore 1.6.321.6.32
CVE-2026-41496GHSA-rg3h-x3jw-7jm5, PYSEC-2026-2923, PYSEC-2026-2952PraisonAI: SQL Injection via unvalidated `table_prefix` in 9 conversation store backends (incomplete fix for CVE-2026-40315)High severityBefore 1.6.81.6.8
CVE-2026-40289GHSA-8x8f-54wf-vv92, PYSEC-2026-467, PYSEC-2026-485PraisonAI Browser Server allows unauthenticated WebSocket clients to hijack connected extension sessionsCritical severityBefore 1.5.1401.5.140
CVE-2026-40288GHSA-vc46-vw85-3wvm, PYSEC-2026-477, PYSEC-2026-488PraisonAI has critical RCE via `type: job` workflow YAMLCritical severityBefore 1.5.1401.5.140
CVE-2026-40287GHSA-g985-wjh9-qxxc, PYSEC-2026-2914, PYSEC-2026-2947PraisonAI Vulnerable to RCE via Automatic tools.py ImportHigh severityBefore 1.5.1401.5.140
CVE-2026-56074GHSA-ffp3-3562-8cv3, PYSEC-2026-2946PraisonAI: Coarse-Grained Tool Approval Cache Bypasses Per-Invocation Consent for Shell CommandsModerate severityBefore 4.5.1284.5.128
CVE-2026-40160GHSA-qq9r-63f6-v542, PYSEC-2026-2951PraisonAIAgents: SSRF via unvalidated URL in `web_crawl` httpx fallbackHigh severityFrom 0.13.23 before 1.5.1281.5.128
GHSA-x462-jjpc-q4q4PraisonAI: Cross-Origin Agent Execution via Hardcoded Wildcard CORS and Missing Authentication on AGUI EndpointHigh severityBefore 4.5.1284.5.128
CVE-2026-40152GHSA-7j2f-xc8p-fjmq, PYSEC-2026-2944PraisonAIAgents: Path Traversal via Unvalidated Glob Pattern in list_files Bypasses Workspace BoundaryModerate severityBefore 1.5.1281.5.128
CVE-2026-40153GHSA-v8g7-9q6v-p3x8, PYSEC-2026-2953PraisonAIAgents: Environment Variable Secret Exfiltration via os.path.expandvars() Bypassing shell=False in Shell ToolHigh severityBefore 1.5.1281.5.128
CVE-2026-40150GHSA-8f4v-xfm9-3244, PYSEC-2026-2945PraisonAIAgents has SSRF and Local File Read via Unvalidated URLs in web_crawl ToolHigh severityBefore 1.5.1281.5.128
CVE-2026-40117GHSA-grrg-5cg9-58pf, PYSEC-2026-2949PraisonAIAgents: Arbitrary File Read via read_skill_file Missing Workspace Boundary and Approval GateModerate severityBefore 1.5.1281.5.128
CVE-2026-40111GHSA-v7px-3835-7gjx, PYSEC-2026-487PraisonAIAgents has an OS Command Injection via shell=True in Memory Hooks Executor (memory/hooks.py)Critical severityBefore 1.5.1281.5.128
CVE-2026-56078GHSA-766v-q9x3-g744, PYSEC-2026-2943PraisonAI has Memory State Leakage and Path Traversal in MultiAgent Context HandlingModerate severityBefore 1.5.1151.5.115
CVE-2026-39888GHSA-qf73-2hrx-xprp, PYSEC-2026-486PraisonAI has sandbox escape via exception frame traversal in `execute_code` (subprocess mode)Critical severityBefore 1.5.1151.5.115
CVE-2026-34954GHSA-44c2-3rw4-5gvh, PYSEC-2026-2940PraisonAI Has SSRF in FileTools.download_file() via Unvalidated URLHigh severityBefore 1.5.951.5.95
CVE-2026-34937GHSA-w37c-qqfp-c67f, PYSEC-2026-2954PraisonAI: Shell Injection in run_python() via Unescaped $() SubstitutionHigh severityBefore 1.5.901.5.90
CVE-2026-34938GHSA-6vh2-h83c-9294, PYSEC-2026-484PraisonAI: Python Sandbox Escape via str Subclass startswith() Override in execute_codeCritical severityBefore 1.5.901.5.90

Other PyPI packages

Sources
  • Package advisories from the GitHub Advisory Database (CC BY 4.0). Version ranges are reworded as sentences (source). Licence: CC BY 4.0
  • Package advisories from the PyPA advisory database (CC BY 4.0). Version ranges are reworded as sentences (source). Licence: CC BY 4.0