← Back to Scanner

coding-agent Security Audit Report

🔭 Continuously monitored by ClawSecure Watchtower
Source:
SHA-256:

coding-agent is an AI agent skill, created by steipete and published at openclaw/skills. ClawSecure audited coding-agent across 2 files through the 3-Layer Audit Protocol covering all ten OWASP ASI Top 10 categories, assigning a security score of 85/100 (Safe). The 4 findings concentrate in Command Injection, Permissions Manifest and Policy Violation, including Variable $SCRATCH (line 2) flows to `bash` at line 5. and Variable $REVIEW_DIR (line 5) flows to `bash` at line 8.. None were rated high or critical severity.

Is coding-agent safe?

ClawSecure audited coding-agent and assigned a security score of 85/100 (Safe), identifying 4 findings across Command Injection and Permissions Manifest. Review the findings below before installing.

What did ClawSecure find in coding-agent?

ClawSecure identified 4 findings in coding-agent, concentrated in Command Injection, Permissions Manifest and Policy Violation. The most severe include Variable $SCRATCH (line 2) flows to `bash` at line 5. and Variable $REVIEW_DIR (line 5) flows to `bash` at line 8..

How was coding-agent audited?

ClawSecure ran coding-agent through its 3-Layer Audit Protocol with full OWASP ASI Top 10 coverage, scanning 2 files from openclaw/skills.

What does a score of 85 mean?

ClawSecure assigned coding-agent a security score of 85/100, placing it in the Safe range. Scores of 80 or above qualify for ClawSecure Verified status; the 4 findings detected are lower-severity. ClawSecure derives this score with a weighted deduction model (critical -20, high -10, medium -5, low -2 from a base of 100).

Audit Findings for coding-agent

ClawSecure detected 4 security findings in coding-agent, spanning Command Injection, Permissions Manifest and Policy Violation.

Each finding is expandable in the interactive list below.

3-Layer Audit Protocol

Security Recommendations for coding-agent

Harden command execution
coding-agent constructs or runs system commands. Validate that commands are built only from trusted inputs, never pass user-controlled strings directly to a shell, and restrict execution to an allow-list of expected commands.
Add a config.json permissions manifest
A config.json file declares what an agent component can access: file system, network, shell execution and more. Without it, users have no visibility into what the component can do before installing. This is the single most impactful security improvement for any AI agent skill.
Resolve policy violations
coding-agent trips ClawSecure policy checks. Review each flagged pattern against your security policy and remediate or document an accepted exception before production use.
Pin dependencies to exact versions
Unpinned dependencies allow supply-chain attacks where a compromised version is pulled in automatically. Use exact version numbers in package.json (for example 1.2.3 instead of ^1.2.3) to keep unauthorized code out of your dependency tree. ClawSecure checks every dependency against known CVE databases.

Related Security Research

Why Generic Scanners Fail at AI Agent Security→Understanding Our 3-Layer Audit Protocol→How to Secure an MCP Server: The 2026 Guide→

Related AI Agent Security Audits

@penpot/mcpScore 75/100shodan-skillScore 85/100everything-claude-codeScore 90/100ibkr-cliScore 95/100claude-code-openclawScore 85/100

Scanned on March 16, 2026. coding-agent is one of thousands of agents audited by ClawSecure from the community-curated awesome-openclaw-skills list and the openclaw/skills repository.

Start Your Free Scan