error-recovery

Classify workflow failures and determine automatic recovery versus manual intervention.

Updated Feb 14, 2026
One-click install
npx skills add https://github.com/DYAI2025/MOD_shark-engine --skill error-recovery-dyai2025
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: error-recovery
Source: https://github.com/DYAI2025/MOD_shark-engine/tree/main/sharkengine/.claude/skills/error-recovery
Command: npx skills add https://github.com/DYAI2025/MOD_shark-engine --skill error-recovery-dyai2025

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Classify workflow failures into actionable categories and determine if auto-fix is possible or manual intervention is required.

Core Features & Use Cases

  • Classifies failures (CI, security, tests, builds, deployments) to guide recovery decisions
  • Applies automatic recovery strategies with a cap of three retry attempts
  • Provides escalation prompts and structured remediation guidance for manual intervention

Quick Start

Load the failure state, classify the cause, and apply auto-recovery steps if safe.

Frequently Asked Questions about error-recovery

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

FAQPage Schema
How do I automate workflow failure recovery when CI or deployment pipelines return a FAILED status?

Workflow failure recovery is automated by classifying the failure cause and applying auto-fix strategies with up to three retry attempts. If the failure classification determines manual intervention is required, it escalates with structured remediation guidance.

What is the best way to classify CI pipeline failures and decide between automatic retries or manual intervention?

Classifying CI pipeline failures involves using a failure classification tree to categorize the error and determine auto-recoverability. The system evaluates the failure state to decide if safe automatic retries are possible or if manual remediation is needed.

How does auto-recovery handle quality gates that block deployment?

Auto-recovery handles blocked deployments by identifying the quality gate failure and applying targeted automatic recovery strategies. It caps these efforts at three retry attempts to prevent infinite loops before escalating the issue for manual intervention.

When should I use an automated error-recovery strategy for sprint implementation failures?

Use an automated error-recovery strategy when sprint or feature implementations fail and require immediate classification. It determines if the workflow phase failure is safely auto-recoverable, logging outcomes and escalation paths for any unresolved manual interventions.

Does automated retry logic work for all types of build and test failures?

Automated retry logic does not work for all build and test failures, as it depends on the failure classification tree. It only applies auto-fix strategies for safely recoverable categories, capping at three attempts before transitioning to manual intervention.

What limitations exist when applying automatic recovery strategies to failed workflows?

Limitations of automatic recovery include a strict cap of three retry attempts for safe auto-fix strategies. Failures classified outside the auto-recoverable categories immediately require manual intervention and structured escalation prompts instead of automated retries.