openclaw - 💡(How to fix) Fix 2026.5.2: security audit recommends tools.exec.strictInlineEval but config schema rejects tools.exec [2 comments, 3 participants]

Official PRs (…)
ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

Helpful · Quick feedback

Loading…
GitHub stats
openclaw/openclaw#77105Fetched 2026-05-05 05:52:07
View on GitHub
Comments
2
Participants
3
Timeline
3
Reactions
2
Author
Timeline (top)
commented ×2closed ×1

openclaw security audit --deep warns that interpreter allowlist entries should set tools.exec.strictInlineEval=true, but adding that exact config key causes the same audit/config validator to print an invalid-config diagnostic: tools.exec: Unrecognized key: "strictInlineEval" / tools.exec rejected under current config shape.

This makes the warning non-actionable for operators on 2026.5.2.

Root Cause

I rolled the local config back because keeping an invalid key is worse than the warning.

Code Example

tools.exec.allowlist_interpreter_without_strict_inline_eval Interpreter allowlist entries are missing strictInlineEval hardening
  Interpreter/runtime allowlist entries were found without strictInlineEval enabled:
- agents.sentinel.allowlist: /usr/bin/find
  Fix: Set tools.exec.strictInlineEval=true (or per-agent tools.exec.strictInlineEval=true) when allowlisting interpreters like python, node, ruby, perl, php, lua, or osascript.

---

"tools": {
  "exec": {
    "strictInlineEval": true
  }
}

---

Invalid config at ~/.openclaw/openclaw.json:
- tools.exec: Unrecognized key: "strictInlineEval"
RAW_BUFFERClick to expand / collapse

Summary

openclaw security audit --deep warns that interpreter allowlist entries should set tools.exec.strictInlineEval=true, but adding that exact config key causes the same audit/config validator to print an invalid-config diagnostic: tools.exec: Unrecognized key: "strictInlineEval" / tools.exec rejected under current config shape.

This makes the warning non-actionable for operators on 2026.5.2.

Environment

  • OpenClaw 2026.5.2 (8b2a6e5)
  • macOS 12.7.6 x64
  • Node 24.14.0
  • Gateway LaunchAgent, loopback 127.0.0.1:18789
  • Personal assistant trust model, Telegram multi-account setup

Audit warning

The audit reports:

tools.exec.allowlist_interpreter_without_strict_inline_eval Interpreter allowlist entries are missing strictInlineEval hardening
  Interpreter/runtime allowlist entries were found without strictInlineEval enabled:
- agents.sentinel.allowlist: /usr/bin/find
  Fix: Set tools.exec.strictInlineEval=true (or per-agent tools.exec.strictInlineEval=true) when allowlisting interpreters like python, node, ruby, perl, php, lua, or osascript.

Attempted fix

Added:

"tools": {
  "exec": {
    "strictInlineEval": true
  }
}

Actual result

Subsequent audit/config validation reports the new key as invalid/unrecognized:

Invalid config at ~/.openclaw/openclaw.json:
- tools.exec: Unrecognized key: "strictInlineEval"

I rolled the local config back because keeping an invalid key is worse than the warning.

Expected

One of these should happen:

  1. The schema accepts tools.exec.strictInlineEval as the audit recommends, or
  2. The audit gives the currently valid per-agent config path for 2026.5.2, or
  3. The warning is suppressed/changed when there is no valid config field available.

Impact

The security audit suggests a hardening action that cannot be applied cleanly on the stable release. Operators either leave the warning unresolved or add invalid config.

No secrets/tokens included.

extent analysis

TL;DR

The tools.exec.strictInlineEval configuration key is not recognized, causing a conflict between the security audit warning and the config validator.

Guidance

  • Review the OpenClaw documentation for the correct configuration key to enable strict inline eval for interpreter allowlist entries.
  • Check the OpenClaw version 2026.5.2 release notes for any known issues or configuration changes related to tools.exec.
  • Consider searching for alternative configuration options or workarounds in the OpenClaw community forums or GitHub issues.
  • Verify the configuration schema for OpenClaw 2026.5.2 to determine the valid configuration keys for tools.exec.

Example

No code snippet is provided as the issue is related to configuration and not code.

Notes

The issue seems to be specific to OpenClaw version 2026.5.2, and the solution may vary depending on the version. It is recommended to check the official documentation and release notes for the correct configuration.

Recommendation

Apply workaround: wait for an update or patch from OpenClaw that addresses the configuration issue, or explore alternative configuration options that achieve the same security hardening goal.

Vote matrix · Quick signals

Works
Did the solution work? Tap to confirm.
Easy Fix
Was it a quick fix?
Time Saver
Did it save you time?
Blocking
Was it severely blocking?
Common Issue
Are others likely hitting this too?
Flaky / Intermittent
Is it intermittent?
Verified / Reproducible
Can you reproduce it reliably?
Loading…

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Back to top recommendations

TRENDING

openclaw - 💡(How to fix) Fix 2026.5.2: security audit recommends tools.exec.strictInlineEval but config schema rejects tools.exec [2 comments, 3 participants]