Can I resume a failed workflow?

Yes, the Run Manager includes a resume-run operation that analyzes the existing state and event log to identify the failure point, allowing the workflow to pick up exactly where it left off.

What is a FABER Run ID?

A Run ID is a unique identifier formatted as {org}/{project}/{uuid} that allows the FABER plugin to track, isolate, and manage specific workflow executions without conflicts.

How does the skill ensure data integrity?

It uses atomic writes for state updates and maintains a sequence-controlled event log. If a state file is corrupted, the reconstruct-state operation can rebuild it from the immutable event history.

Does this support parallel workflow execution?

Yes, by generating unique IDs and isolated directory structures for every run, it prevents multiple processes from sharing or overwriting the same state data.

FABER Run Manager

Name: FABER Run Manager
Author: fractary

byfractary

•

生产力与工作流

Manages the complete lifecycle of FABER workflow runs, providing robust state tracking, event logging, and execution recovery.

关于

The Run Manager skill is a core component of the FABER plugin system, designed to handle the orchestration and auditing of complex AI workflows. It ensures every execution has a unique identity, maintains an atomic record of state changes through an event-driven architecture, and enables advanced features like resuming from failures or rerunning specific steps with modified parameters. This is particularly useful for developers building long-running autonomous processes that require high reliability, parallel execution capability, and a clear audit trail for debugging and compliance.

主要功能

Event-Driven State Tracking: Emits sequence-controlled events for a complete audit trail of workflow progress.
Execution Recovery: Supports resuming interrupted runs from the exact point of failure using saved state files.
Unique Run ID Generation: Creates organization and project-specific identifiers for conflict-free parallel execution.
Rerun & Parameter Override: Enables re-executing workflows with modified parameters while preserving the original context.
State Reconstruction: Rebuilds corrupted state files from event history to ensure data integrity and system reliability.
2 GitHub stars

使用场景

Recovering complex AI-driven deployment workflows after intermittent network or API failures.
Managing parallel experimentation runs with distinct parameter sets and isolated run environments.
Orchestrating multi-step autonomous coding tasks with full observability and sequence logging.

关于

主要功能

Event-Driven State Tracking: Emits sequence-controlled events for a complete audit trail of workflow progress.
Execution Recovery: Supports resuming interrupted runs from the exact point of failure using saved state files.
Unique Run ID Generation: Creates organization and project-specific identifiers for conflict-free parallel execution.
Rerun & Parameter Override: Enables re-executing workflows with modified parameters while preserving the original context.
State Reconstruction: Rebuilds corrupted state files from event history to ensure data integrity and system reliability.
2 GitHub stars

使用场景

Recovering complex AI-driven deployment workflows after intermittent network or API failures.
Managing parallel experimentation runs with distinct parameter sets and isolated run environments.
Orchestrating multi-step autonomous coding tasks with full observability and sequence logging.