{"answers":{"affected":"3 open source packages ship an affected version, including praisonai. It affects 1 product.","exploited":"It is not confirmed as exploited in the wild, but public exploit code for CVE-2026-39891 exists.","fix":"Upgrade praisonai to 4.5.115 or later.","what":"CVE-2026-39891 is a high-severity code injection vulnerability in MervinPraison PraisonAI, published on 8 Apr 2026. Any signed-in user can exploit it over the network, with no action from the victim."},"fix_prompt":"Check this codebase for CVE-2026-39891 and fix it if it is affected.\n\nWhat it is: CVE-2026-39891 is a high-severity code injection vulnerability in MervinPraison PraisonAI, published on 8 Apr 2026. Any signed-in user can exploit it over the network, with no action from the victim.\nExploitation: It is not confirmed as exploited in the wild, but public exploit code for CVE-2026-39891 exists.\n\nAffected packages and fixed versions:\n- praisonai (PyPI): affected all versions; fixed in 4.5.115\n- praisonai (PyPI): affected all versions; fixed in 4.5.115\n- praisonai (PyPI): affected all versions; fixed in 4.5.115\n\nTask:\n1. Search the dependency manifests and lockfiles (for example package.json, package-lock.json, requirements.txt, poetry.lock, uv.lock, pom.xml, build.gradle, go.mod, Cargo.lock, Gemfile.lock) and any Dockerfiles for the affected packages, including transitive dependencies.\n2. If an affected version is in use, upgrade it to the fixed version for its release line with the project's package manager, so the lockfile is regenerated. Prefer the smallest upgrade that fixes it.\n3. If there is no fix, or the upgrade would break the build, do not guess: explain the risk and the safest workaround instead.\n4. Run the tests and summarise what changed and why.\n5. If nothing in the project is affected, say so and change nothing.\n\nReference: https://insights.sbomify.com/CVE-2026-39891/ (Markdown: https://insights.sbomify.com/CVE-2026-39891/index.md)","id":"CVE-2026-39891","markdown":"https://insights.sbomify.com/CVE-2026-39891/index.md","name":"CVE-2026-39891","record":{"aliases":["GHSA-hwg5-x759-7wjg","PYSEC-2026-2917"],"analysis":{"component_type":"library","fix_type":"upgrade","impacts":["code_execution"],"network_reachable":"yes","non_default_config":false,"product_category":"Python AI agent library","requires_auth":"low_privilege","summary":"Attackers can execute arbitrary code via template injection in PraisonAI agent tools when user input is not escaped.","user_interaction":"not_required","vuln_class":"code_injection"},"analysis_generated_by":"qwen3.8-flash-next","assigner":"GitHub_M","cpe":[{"product":"praisonai","vendor":"praison"}],"cvss":{"score":8.8,"severity":"high","source":"cna","vector":"CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H","version":"3.1"},"cwes":[{"id":"CWE-94","name":"Improper Control of Generation of Code"}],"description":"PraisonAI is a multi-agent teams system. Prior to 4.5.115, the create_agent_centric_tools() function returns tools (like acp_create_file) that process file content using template rendering. When user input from agent.start() is passed directly into these tools without escaping, template expressions in the input are executed rather than treated as literal text. This vulnerability is fixed in 4.5.115.","ecosystems":["PyPI"],"exploits":{"exploitdb":0,"first_seen":null,"github_poc":0,"max_poc_stars":null,"metasploit":0,"nuclei":0,"refs":[],"tier":"weaponised"},"has_fix":true,"headline":"PraisonAI has a Template Injection in Agent Tool Definitions","id":"CVE-2026-39891","indexable":true,"kev":null,"kind":"cve","links":{"ecosystem":"PyPI","product":null,"products":[],"type":"code_injection","year":"2026"},"n_packages":3,"n_references":1,"next":{"headline":"cryptography has a buffer overflow if non-contiguous buffers were passed to APIs","id":"CVE-2026-39892"},"packages":[{"ecosystem":"PyPI","fixed":["4.5.115"],"introduced":["0"],"last_affected":[],"package":"praisonai","purl":null,"source":"GHSA-hwg5-x759-7wjg"},{"ecosystem":"PyPI","fixed":["4.5.115"],"introduced":["0"],"last_affected":[],"package":"praisonai","purl":"pkg:pypi/praisonai","source":"GHSA-hwg5-x759-7wjg"},{"ecosystem":"PyPI","fixed":["4.5.115"],"introduced":["0"],"last_affected":[],"package":"praisonai","purl":"pkg:pypi/praisonai","source":"PYSEC-2026-2917"}],"prev":{"headline":"PraisonAI Affected by Remote Code Execution via YAML Deserialization in Agent Definition Loading","id":"CVE-2026-39890"},"products":["PraisonAI"],"published":"2026-04-08","related":[{"cvss_score":10,"headline":"PraisonAIAgents has a sandbox escape via exception frame traversal in `execute_code` (subprocess mode)","id":"CVE-2026-39888","kev_date_added":null,"published":"2026-04-08"},{"cvss_score":9.8,"headline":"Incomplete fix for CVE-2026-34935: Command Injection in MervinPraison/PraisonAI","id":"CVE-2026-41497","kev_date_added":null,"published":"2026-05-08"},{"cvss_score":9.8,"headline":"PraisonAI: Critical RCE via `type: job` workflow YAML","id":"CVE-2026-40288","kev_date_added":null,"published":"2026-04-14"},{"cvss_score":9.8,"headline":"PraisonAI Affected by Remote Code Execution via YAML Deserialization in Agent Definition Loading","id":"CVE-2026-39890","kev_date_added":null,"published":"2026-04-08"},{"cvss_score":9.8,"headline":"PraisonAI: Second-Order SQL Injection in `get_all_user_threads`","id":"CVE-2026-34934","kev_date_added":null,"published":"2026-04-03"},{"cvss_score":9.8,"headline":"PraisonAI: OS Command Injection in MCPHandler.parse_mcp_command()","id":"CVE-2026-34935","kev_date_added":null,"published":"2026-04-03"},{"cvss_score":9.7,"headline":"Improper Neutralization of Special Elements used in an OS Command ('OS Command Injection') in praisonai","id":"CVE-2026-40088","kev_date_added":null,"published":"2026-04-09"}],"sources":[{"class":"attribution","licence":"CVE Terms of Use (MITRE)","name":"CVE List","prefix":"CVE"},{"class":"attribution","licence":"CC BY 4.0","name":"GitHub Advisory Database","prefix":"GHSA"},{"class":"attribution","licence":"CC BY 4.0","name":"PyPA Advisory Database","prefix":"PYSEC"},{"class":"attribution","licence":"US public information, NVD API terms","name":"NVD"},{"class":"attribution","licence":"CWE Terms of Use","name":"CWE (MITRE)"},{"class":"attribution","licence":"FIRST CVSS licence","name":"CVSS (FIRST)"}],"ssvc":{"automatable":"no","exploitation":"poc","technical_impact":"total"},"state":"PUBLISHED","updated":"2026-04-09","vendors":["MervinPraison"]},"summary":"Attackers can execute arbitrary code via template injection in PraisonAI agent tools when user input is not escaped.","url":"https://insights.sbomify.com/CVE-2026-39891/"}