rad-log-error

Appends structured error entries to a project error log when pipeline execution fails.

Updated May 24, 2026
One-click install
npx skills add https://github.com/MetalHexx/rad-orc-marketplace --skill rad-log-error-metalhexx
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: rad-log-error
Source: https://github.com/MetalHexx/rad-orc-marketplace/tree/main/claude-plugin/skills/rad-log-error
Command: npx skills add https://github.com/MetalHexx/rad-orc-marketplace --skill rad-log-error-metalhexx

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? When a radorch pipeline execution fails, agent output is invalid, or manual intervention is required, there is no persistent record of what went wrong. This Skill creates and maintains a structured, append-only Markdown error log so every failure is captured with context for later diagnosis and recovery. ## Core Features & Use Cases - Structured Error Entries: Appends entries with timestamp, pipeline event, action, severity, phase, and task index using a fixed table-and-section template. - Severity Classification: Classifies failures as critical, high, medium, or low using a defined criteria guide. - Append-Only Log Management: Creates the error log from a bundled template on first failure, tracks entry_count in frontmatter, and never modifies existing entries. - Use Case: When the radorch pipeline signal envelope returns ok: false, the orchestrator invokes this Skill to record the raw JSON output, root cause, and workaround in {PROJECT}-ERROR-LOG.md. ## Quick Start Log the current pipeline failure, including its raw JSON envelope and severity, as a new entry in the project error log.

Frequently Asked Questions about rad-log-error

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

FAQPage Schema
How do I log a pipeline failure in a persistent error log?▼

Invoke this Skill when the radorch pipeline signal envelope returns ok: false. It creates or updates {PROJECT}-ERROR-LOG.md, appending a structured entry with timestamp, event, severity, root cause, and workaround.

What information should a pipeline error log entry contain?▼

Each entry records the entry number, ISO-8601 timestamp, pipeline event and action, severity level, phase and task index, symptom description, raw JSON pipeline output, root cause, and any workaround applied.

How is error severity classified for pipeline failures?▼

Severity uses four levels: critical blocks all execution, high produces incorrect state, medium degrades behavior with workarounds, and low is cosmetic. The Skill provides a classification guide with concrete examples for each level.

Can I edit or delete existing entries in the error log?▼

No. The error log is strictly append-only. Existing entries are never modified or deleted; only new entries are appended and the frontmatter entry_count and last_updated fields are updated.

When should this error logging skill be triggered?▼

Trigger it whenever the pipeline signal envelope returns ok: false, when an agent produces invalid output, or when manual intervention is needed during pipeline execution. Logging failures is near-mandatory for a persistent record.