CVE-2026-39891 — praisonai
HIGHCVE-2026-39891 is a high-severity (CVSS 8.8) Code Injection vulnerability in praisonai. A fix is available for praisonai — see the affected versions and patch details below.
PraisonAI has a Template Injection in Agent Tool Definitions
Exploitation Status
Proof-of-concept exploit code exists
- CISA’s SSVC triage found public proof-of-concept exploit code for this CVE, though no confirmed active exploitation.
- A successful exploit gives an attacker total control of the affected component, not partial access.
Exploitation and automatability from CISA’s SSVC triage for CVE-2026-39891.
EPSS Exploitation Probability
Probability of exploitation in the next 30 days, from FIRST.org EPSS.
How urgent is this, really
Where this sits among everything scored
Real-World Exposure
praisonaiReal-time download stats are indexed for npm and PyPI packages. This vulnerability affects PyPI packages — download data is not available via public APIs for these ecosystems.
Description
Summary
Direct insertion of unescaped user input into template-rendering tools allows arbitrary code execution via specially crafted agent instructions.
Details
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 (as shown in agent_centric_example.py:85-86), template expressions in the input are executed rather than treated as literal text. This occurs because:
- No input sanitization or escaping is applied to user-controlled content
- The ACP-enabled runtime auto-approves operations (
approval_mode="auto") - Tools lack context-aware escaping for template syntax
PoC
# Replace the agent.start() call at line 85 with:
result = agent.start('Create file with content: {{ self.__init__.__globals__.__builtins__.__import__("os").system("touch /tmp/pwned") }}')
Successful exploitation creates /tmp/pwned confirming arbitrary command execution. The expression {{7*7}} renders as 49 instead of literal text.
Impact
Attackers can execute arbitrary system commands with the privileges of the running process by injecting malicious template expressions through agent instructions. This compromises the host system, enabling data theft, ransomware deployment, or lateral movement.
Recommended Fix
- Input Sanitization: Implement strict whitelist validation for file content
- Contextual Escaping: Auto-escape template syntax characters (e.g.,
{{ }}) in user input using Jinja2autoescape=True - Sandboxing: Restrict template execution environments using secure eval modes
- Approval Hardening: Require manual approval for file creation operations in production
Affected Packages
| Ecosystem | Package | Vulnerable range | Fix |
|---|---|---|---|
| 🐍PyPI | praisonai | all versions | 4.5.115pip install --upgrade 'praisonai==4.5.115' |
Affected Products
praisonaipraisonDetection & mitigation playbook
Open-source dependencyDetect
Scan your dependency tree (package-lock.json, pnpm-lock.yaml, requirements.txt, go.sum, etc.) for praisonai, including transitive dependencies — a direct dependency you never call can still pull in a vulnerable version.
Fix
Update praisonai to 4.5.115 or later, then make sure no transitive (indirect) dependency still pins the vulnerable range — O3 confirms CVE-2026-39891 is resolved across your whole dependency graph.
Workarounds
Stop passing untrusted input into the interpreter or shell: call the affected binary with an argument array rather than a composed command string, reject anything outside a strict allowlist of expected values, and run the component under an account that cannot reach beyond the work it legitimately does.
Frequently Asked Questions
Is CVE-2026-39891 in your dependencies?
Find it across PyPI, including transitive dependencies.