// Security Advisory Triage

PraisonAI agents package executes scripts without path validation

GitHub advisories — pip · published 2026-10-09 13:20 UTC
ACTION NEEDED PATCH_NOW SECURITYAPI
WHAT CHANGEDGitHub published CVE-2026-61443 on 8 October 2026, a high-severity vulnerability in praisonaiagents. The SkillTools component executes scripts without validating file paths, allowing arbitrary script execution. Version 1.6.78 patches the issue.
WHY IT MATTERSAn attacker who can influence script paths passed to SkillTools can execute arbitrary code on the system running the agent. This bypasses intended containment and gives full process privileges to untrusted input.
WHAT TO DOCheck your dependency tree for praisonaiagents. If present and below 1.6.78, update to 1.6.78 or later this week. If you cannot update immediately, audit all code paths that pass file arguments to SkillTools and ensure they reject user-controlled input.
Model-reported confidence 85% (not measured accuracy) Primary source ↗ written by anthropic/claude-sonnet-4.5
Show the reasoning — raw diff, triage verdict, confidence
RAW DIFF
+ghsa_id: GHSA-c44f-37qr-gw3f
+summary: PraisonAI: SkillTools Executes Scripts Without Path Containment Validation
+severity: high
+published_at: 2026-10-08T16:49:23Z
+cve_id: CVE-2026-61443
+vulnerabilities[].package.name: praisonaiagents
+vulnerabilities[].first_patched_version: 1.6.78
HOW IT WAS JUDGED
Triage signals below are model-reported assertions, preserved verbatim; they are not independent arithmetic verification.
triage verdictMATERIAL at 0.78, needed 0.5
triage reasoningPraisonAI is an AI agent framework in the dependency tree of services using agentic patterns. Script execution without path validation is exploitable.
signalsAI agent framework package · arbitrary script execution vulnerability · path containment bypass · high severity · patched version available · new advisory entry
suppression—
triage modelanthropic/claude-haiku-4.5 ($0.00000)
writer modelanthropic/claude-sonnet-4.5 ($0.00000)