evidence-url-verifier

What This Skill Does

The Evidence URL Verifier is a mission-critical utility for OpenClaw agents designed to validate the authenticity and accessibility of external resources provided as proof of work. In automated environments, "placeholder" URLs, broken links, and empty artifacts can lead to silent task failures or falsified reporting. This skill automates the verification process by performing deep head requests to ensure resources resolve, inspecting content-types to prevent mismatches, and parsing response bodies for common placeholder text like "lorem ipsum" or "example.com". By integrating this into your agent's workflow, you ensure that any evidence cited in a report is verifiable, substantive, and accessible to human auditors.

Installation

To integrate this skill into your local environment, ensure you have the OpenClaw agent installed. Run the following command in your terminal:

clawhub install openclaw/skills/skills/dalomeve/evidence-url-verifier

Once installed, the agent will have access to the Test-EvidenceUrl function and the associated content inspection logic defined in the skill manifest.

Use Cases

• Automated Reporting: Verify that links included in progress reports are active before submitting to management.
• Quality Assurance: Check that generated artifacts (e.g., screenshots or logs) are not zero-byte files.
• Content Auditing: Ensure that marketing or documentation links point to final production pages rather than development placeholders.
• Compliance Checks: Validate that external links adhere to expected content types (e.g., ensuring a PDF link actually returns application/pdf).

Example Prompts

1. "Verify the evidence URL provided in the latest task update to ensure it isn't a placeholder."
2. "Check all external links in the current report for 200 OK responses and ensure no 'lorem ipsum' content is found."
3. "Before marking this audit as complete, run the evidence-url-verifier on all attached file paths and external URLs."

Tips & Limitations

• Rate Limiting: To avoid IP blacklisting or triggering WAFs, this skill enforces a strict 1 request per second limit. Do not bypass this setting.
• Privacy: Avoid passing URLs containing API keys or private tokens, as the skill may log request headers. The skill is designed to ignore private contents but always sanitize inputs.
• Scope: This tool focuses on public or accessible file paths. It cannot bypass authentication barriers or VPN requirements.