In this article
Experimental/Mural/Mural Log Hygiene
| Field | Value |
|---|---|
| Kind | instruction |
| Source | .github/instructions/experimental/mural/mural-log-hygiene.instructions.md |
| Invocation | Applied automatically to **/.copilot-tracking/mural/**, **/.github/skills/experimental/mural/**, **/.github/agents/design-thinking/dt-coach.agent.md, **/.github/agents/rai-planning/rai-planner.agent.md, **/.github/agents/project-planning/ux-ui-designer.agent.md, **/.github/instructions/experimental/mural/** |
| Interactive | No |
What it does
Operator log-hygiene contract for Mural customizations: never echo raw URLs, Azure SAS query strings, OAuth tokens, or Authorization headers; the skill _redact() is a defense-in-depth backstop, not a license to log.
When to use it
Apply these rules whenever Mural commands, customizations, diagnostics, or network evidence could expose URLs, headers, tokens, authorization codes, or Azure SAS values. Use only redacted evidence outside the terminal and treat the skill's redaction helper as defense in depth, not permission to log raw traffic.
Example usage
Provide a concrete example that shows the asset in action, including representative input and the resulting output.