OpenClaw Code Search

Security checks across static analysis, malware telemetry, and agentic risk

Overview

This is a coherent read-only code search skill; users should mainly notice that it can reveal local file contents to the agent and relies on external CLI tools.

This appears safe for local code search. Install or verify the required tools from trusted sources, use it on specific project directories rather than broad private paths, and remember that returned file contents may include secrets or untrusted instructions.

SkillSpector

By NVIDIA

SkillSpector has not run for this release. Legacy ClawScan findings remain available under Risk analysis.

Static analysis

No static analysis findings were reported for this release.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Risk analysis

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

#
ASI02: Tool Misuse and Exploitation
Low
What this means

If invoked on a sensitive directory, search results could show private filenames or matched file contents in the agent conversation.

Why it was flagged

The script runs local search tools against caller-provided patterns and paths. Arguments are passed safely as arrays and the operation is read-only, but broad paths could expose more local content than intended.

Skill content
rg_args+=("$pattern" "$search_path") ... raw_output=$(timeout 10 rg "${rg_args[@]}" 2>/dev/null || true)
Recommendation

Use the skill on intended project folders, avoid broad paths like home or root directories, and review searches involving secrets or private documents.

#
ASI04: Agentic Supply Chain Vulnerabilities
Low
What this means

Installing dependencies from unverified downloads could expose the environment to supply-chain risk if the source or download is compromised.

Why it was flagged

The dependency guidance includes manual remote binary downloads and installation to a system path without checksum verification. This is not automatic skill behavior and is purpose-aligned, but users should verify provenance.

Skill content
curl -LO https://github.com/BurntSushi/ripgrep/releases/download/14.1.1/ripgrep-14.1.1-x86_64-unknown-linux-musl.tar.gz
tar xzf ripgrep-*.tar.gz && cp ripgrep-*/rg /usr/local/bin/
Recommendation

Prefer trusted package-manager installations where possible, or verify release signatures/checksums before manually installing rg, fd, or tree.

#
ASI06: Memory and Context Poisoning
Low
What this means

Matched code or comments could reveal sensitive strings or attempt to influence the agent if interpreted as instructions.

Why it was flagged

The grep command returns matched file lines into the agent context. This is expected for code search, but repository text may contain secrets or prompt-like instructions that should be treated as untrusted data.

Skill content
printf "  %s: %s\n", line_num, line_text
Recommendation

Treat search output as data, not authority; avoid searching secret-heavy paths unless necessary and review sensitive matches before sharing them further.