{"schemaVersion":1,"url":"https://cve.openhack.com/cve/CVE-2026-57142","id":"CVE-2026-57142","title":"PraisonAI dangerous-tool policy bypass in recipe execution","description":"PraisonAI is a multi-agent teams system. From version 4.5.87 to before version 4.6.61, PraisonAI recipe execution has a dangerous-tool policy that is supposed to block default-denied tools unless the caller explicitly passes allow_dangerous_tools=True. That policy only checks tools declared in TEMPLATE.yaml requires.tools. For steps-based recipes, the actual execution path loads workflow.yaml with","descriptionLong":"PraisonAI is a multi-agent teams system. From version 4.5.87 to before version 4.6.61, PraisonAI recipe execution has a dangerous-tool policy that is supposed to block default-denied tools unless the caller explicitly passes allow_dangerous_tools=True. That policy only checks tools declared in TEMPLATE.yaml requires.tools. For steps-based recipes, the actual execution path loads workflow.yaml with YAMLWorkflowParser. That parser resolves agent-level tools: declarations and preserves top-level approve:. Workflow.start() then installs those YAML-approved tools into the approval context. As a result, an untrusted recipe can omit execute_command from TEMPLATE.yaml requires.tools, declare it in workflow.yaml agents.*.tools, and add top-level approve: [execute_command]. The caller did not set allow_dangerous_tools=True, but the recipe policy allows the recipe and the workflow approval path self-approves the critical shell tool. This issue has been patched in version 4.6.61.","published":"2026-10-23T12:07:46.547Z","updated":"2026-10-29T14:17:57.597Z","reserved":"2026-06-24T01:45:48.697Z","status":"Deferred","rejected":false,"assigner":"GitHub_M","metric":{"version":"3.1","score":7.8,"vector":"CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H","rating":"high","source":"GitHub_M"},"metrics":[{"version":"3.1","score":7.8,"vector":"CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H","rating":"high","source":"GitHub_M"},{"version":"3.1","score":7.8,"vector":"CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H","rating":"high","source":"github.com"}],"rating":"high","products":["MervinPraison PraisonAI"],"affectedProducts":[{"vendor":"MervinPraison","product":"PraisonAI","defaultStatus":null,"versions":[{"version":">= 4.5.87, < 4.6.61","status":"affected"}],"platforms":[]}],"packages":[],"cwes":["CWE-78","CWE-94","CWE-863"],"aliases":[],"references":[{"url":"https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-7qw2-w5rc-37x2","tags":["x_refsource_CONFIRM","exploit"]}],"patchReferenced":false,"fixAvailable":false,"exploitReferenced":true,"kev":false,"kevAdded":null,"kevDue":null,"kevEntry":null,"ssvc":{"exploitation":"poc","automatable":"no","technicalImpact":"total"},"epss":0.0024,"epssPercentile":0.15342,"sources":[{"name":"CVE List","url":"https://www.cve.org/CVERecord?id=CVE-2026-57142","ok":true},{"name":"NVD","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-57142","ok":true,"note":"Deferred"},{"name":"OSV","url":"https://osv.dev/vulnerability/CVE-2026-57142","ok":false,"note":"no open-source package data"},{"name":"CISA KEV","url":"https://www.cisa.gov/known-exploited-vulnerabilities-catalog","ok":true,"note":"not listed"},{"name":"EPSS","url":"https://api.first.org/data/v1/epss?cve=CVE-2026-57142","ok":true,"note":"scored 2026-11-19"}],"aiTitle":true,"tech":[],"critical":[],"attention":null}