Brand Sentiment Optimizer

PassAudited by VirusTotal on May 11, 2026.

Overview

Type: OpenClaw Skill Name: geo-sentiment-optimizer Version: 1.0.0 The `SKILL.md` file instructs the agent to execute a `bash` command (`python scripts/audit_sentiment.py --brand <domain> --output report.md`) that includes user-controlled arguments (`<domain>`, `report.md`). This pattern presents a shell injection vulnerability if the OpenClaw agent does not properly sanitize or quote these inputs before executing the command. While the provided `scripts/audit_sentiment.py` does not actually accept `--brand` or `--output` arguments (it takes `--content`), the instruction itself represents a risky capability that could be exploited if the agent's execution environment is not robust against such inputs, or if a different script were provided that did accept and unsafely use these arguments.

Findings (0)

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.

What this means

A documented command may fail or prompt a user/agent to look for code that was not reviewed with this skill.

Why it was flagged

The skill documents a score_sentiment.py tool, but the provided file manifest/source include only scripts/audit_sentiment.py. This makes the tool catalog incomplete and users should not fetch or substitute unreviewed code to satisfy the missing command.

Skill content
python scripts/score_sentiment.py --content content.md
Recommendation

Use only the included files, or ask the maintainer to provide and register the missing script before running it.

What this means

Users may think the included script audits a website or writes a report, when it only scores supplied text.

Why it was flagged

The actual script accepts only a --content argument, while SKILL.md shows commands using --brand, --output, and --pages. This is a documentation/capability mismatch rather than evidence of malicious behavior.

Skill content
parser.add_argument("--content", required=True)
Recommendation

Treat the script as a simple text scorer unless the maintainer updates the code and documentation to match.