Adversarial Prompting
PassAudited by VirusTotal on May 12, 2026.
Overview
Type: OpenClaw Skill Name: adversarial-prompting Version: 1.0.0 The skill bundle is designed for adversarial problem-solving and includes a Python script (`scripts/export_analysis.py`) to save the agent's analysis output. The script writes a markdown file containing the analysis to the user's home directory (`Path.home()`). This behavior is clearly aligned with the stated purpose of the skill, which is to generate detailed analysis, and does not exhibit any signs of data exfiltration, malicious execution, persistence, or prompt injection with harmful intent. The file writing operation is local and targets a non-sensitive user-owned location.
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.
Using the skill may create a local markdown file as part of normal operation.
This directs the agent to run an included helper script after use. The behavior is disclosed and purpose-aligned, but it is automatic rather than separately confirmed.
After presenting the analysis, automatically export the complete output to a markdown file using `scripts/export_analysis.py`.
If you do not want local files created, tell the agent not to run the export step or delete the generated file afterward.
If the analysis includes private business details, personal information, or secrets, those details may remain in a local file.
The helper saves the complete analysis content to a markdown file in the user's home directory, creating a persistent local record.
filepath = Path.home() / filename ... f.write(content)
Avoid including sensitive information unless you are comfortable storing it locally, and remove the generated markdown file when it is no longer needed.