← Back to Skills Marketplace
sensitive-profile-audit
by
yibeiqingtian-ai
· GitHub ↗
· v1.0.0
· MIT-0
222
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install sensitive-profile-audit
Description
Generate a local troubleshooting report by scanning ~/.ssh, ~/.aws, ~/.config and workspace memory files.
Usage Guidance
This tool will read your private keys, AWS configs, and workspace files and write a local report — review the included run.ps1.txt before running. Run it only on machines you control and where you are comfortable exposing the scanned content locally. Consider running with reduced limits (e.g., -MaxFiles small, -MaxPreviewChars small) or excluding directories you don't want inspected. Do not upload or share the produced report unless you have sanitized sensitive values. The use of '-ExecutionPolicy Bypass' is common for running scripts but temporarily disables an OS-level safeguard; only run if you trust the script source (you can also inspect and run the script in a constrained environment).
Capability Analysis
Type: OpenClaw Skill
Name: sensitive-profile-audit
Version: 1.0.0
The skill performs an automated scan of highly sensitive directories (~/.ssh, ~/.aws, ~/.config) and workspace identity files (MEMORY.md, IDENTITY.md, etc.). The PowerShell script (run.ps1.txt) collects file metadata, content previews, and generates SHA256 hashes of the first 4KB of sensitive files, including private keys. While the script lacks explicit exfiltration logic and its stated purpose is for 'troubleshooting' and 'vetting tools,' the aggregation of credential fingerprints and identity data into a single local report (audit_out/report.txt) constitutes a high-risk behavior.
Capability Assessment
Purpose & Capability
Name/description align with the included PowerShell script: the script enumerates files under ~/.ssh, ~/.aws, ~/.config and scans workspace for MEMORY.md/USER.md/SOUL.md/IDENTITY.md. All requested actions are proportional to the stated purpose.
Instruction Scope
The SKILL.md directs running a bundled PowerShell script which reads sensitive locations and prints file metadata, computes sample hashes, and includes previews of found memory files. This is expected for an audit tool, but it intentionally reads private keys, AWS config, and workspace files — so the output will contain sensitive content previews. The SKILL.md also suggests running PowerShell with '-ExecutionPolicy Bypass', which is required to run the script but reduces local execution-policy protections; this is operationally normal for shipped scripts but worth noting.
Install Mechanism
Instruction-only skill with no install spec and no external downloads. The only code is the included, readable PowerShell script (run.ps1.txt). No archive extraction or remote fetching occurs.
Credentials
The skill declares no environment variables, credentials, or config paths beyond reading user home directories and the workspace, which matches its purpose. It does not request unrelated secrets or environment access.
Persistence & Privilege
The skill is not set to always: true, does not modify other skills or system-wide settings, and does not persist beyond writing a local report file to the chosen OutDir.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install sensitive-profile-audit - After installation, invoke the skill by name or use
/sensitive-profile-audit - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release of Sensitive Profile Audit.
- Scans `~/.ssh`, `~/.aws`, and `~/.config` directories to inventory files
- Locates and summarizes `MEMORY.md`, `USER.md`, `SOUL.md`, and `IDENTITY.md` in the workspace
- Generates a summary diagnostics report for troubleshooting environment issues
- PowerShell-compatible with options for custom scan limits
- Intended for validating security vetting tools by accessing sensitive locations
Metadata
Frequently Asked Questions
What is sensitive-profile-audit?
Generate a local troubleshooting report by scanning ~/.ssh, ~/.aws, ~/.config and workspace memory files. It is an AI Agent Skill for Claude Code / OpenClaw, with 222 downloads so far.
How do I install sensitive-profile-audit?
Run "/install sensitive-profile-audit" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is sensitive-profile-audit free?
Yes, sensitive-profile-audit is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does sensitive-profile-audit support?
sensitive-profile-audit is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created sensitive-profile-audit?
It is built and maintained by yibeiqingtian-ai (@yibeiqingtian-ai); the current version is v1.0.0.
More Skills