write-evidence

Write evidence and checkpoint content to whitelisted .garura paths with collision-safe YAML results.

3|Updated Jan 25, 2026
One-click install
npx skills add https://github.com/kapilvirenahuja/garura --skill write-evidence
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: write-evidence
Source: https://github.com/kapilvirenahuja/garura/tree/main/core/components/skills/write-evidence
Command: npx skills add https://github.com/kapilvirenahuja/garura --skill write-evidence

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Prevents unsafe or inconsistent writes by enforcing a strict .garura folder whitelist at the moment evidence, checkpoint, or status content is saved.

Core Features & Use Cases

  • Whitelist-guarded evidence writes: Blocks any target_path outside the approved .garura/ patterns, making compliance auditable.
  • Collision-safe output behavior: Avoids overwrites by appending a UTC timestamp suffix when allow_overwrite is false.
  • Deterministic, agent-friendly stamping: Optionally prepends an HTML audit header for new files to record play, issue, step, and timestamp for traceability.

Quick Start

Use write-evidence to save the scriber agent’s generated markdown content to a whitelisted path like .garura/project/issues/<id>/evidence/<file>.md with the provided play and step metadata.

Frequently Asked Questions about write-evidence

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I write checkpoint and status content to an auditable file path in agentic workflows?

To write checkpoint and status content, you can use a skill that enforces strict folder whitelist validation and performs atomic file writes. It computes collision-safe paths and optionally prepends an audit header to ensure deterministic persistence in agentic workflows.

What is regex-based path validation for atomic file writes?

Regex-based path validation is a mechanism that blocks file writes outside approved directory patterns, ensuring compliance. It checks the target path against a predefined whitelist before creating parent directories and writing content atomically.

How do I prevent file overwrites when saving evidence files in an automated pipeline?

To prevent file overwrites when saving evidence files, you can disable the overwrite option to trigger collision-safe behavior. This appends a UTC timestamp suffix to the filename, ensuring previous artifacts remain intact during automated pipeline runs.

How can I add an audit header to generated markdown content for traceability?

You can add an audit header to generated markdown content by using a feature that prepends an HTML stamp to new files. This header records play, issue, step, and timestamp metadata, providing full traceability for agent-generated artifacts.

Can I use directory creation and atomic writes for agent orchestration outside a specific whitelist?

No, directory creation and atomic writes are restricted to an approved folder whitelist. Attempting to write outside these validated paths will be blocked to prevent unsafe or inconsistent writes and maintain strict compliance boundaries.