recover-stalled-subagent

Create a WIP commit from a stalled subagent's worktree and remove it.

Updated Mar 13, 2026
One-click install
npx skills add https://github.com/darylmcd/Roslyn-Backed-MCP --skill recover-stalled-subagent
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: recover-stalled-subagent
Source: https://github.com/darylmcd/Roslyn-Backed-MCP/tree/main/.claude/skills/recover-stalled-subagent
Command: npx skills add https://github.com/darylmcd/Roslyn-Backed-MCP --skill recover-stalled-subagent

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The recovery operator tool salvages uncommitted work from a stalled initiative-executor subagent by creating a durable WIP commit on the remediation branch and cleanly removing its worktree, ensuring durable progress is preserved for orchestration.

Core Features & Use Cases

  • Validate that a subagent's worktree is recoverable and bound to remediation/<id>, then stage and commit local changes as a WIP with an informative message.
  • Push the recovered work to origin when needed and emit a STATE_HANDOFF block for downstream planning, enabling retry, defer, or hand-off decisions.
  • Safely remove the worktree and avoid deleting the remediation branch, while providing guardrails to prevent data loss and ensure reproducible recovery outcomes.

Quick Start

Provide the target remediation branch name to begin the recovery process.

Frequently Asked Questions about recover-stalled-subagent

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

FAQPage Schema
How do I recover uncommitted work from a stalled git worktree?

To recover uncommitted work from a stalled git worktree, this tool stages local changes and creates a durable WIP commit on the remediation branch. This salvages your progress safely before the worktree is removed.

What is the best way to handle subagent state handoff during orchestration?

Subagent state handoff during orchestration is handled by emitting a STATE_HANDOFF block if push or removal fails. This leaves the decision to retry or defer to the orchestrator, ensuring downstream planning receives known context.

How do I preserve WIP commits safely when a subagent stalls?

You can preserve WIP commits safely when a subagent stalls by validating preconditions with strict guards and committing local changes to a remediation branch. This approach avoids deleting the branch while ensuring reproducible recovery outcomes.

Does the recovery process delete the remediation branch?

No, the recovery process does not delete the remediation branch. It enforces safety controls to avoid losing work or deleting the wrong branch, focusing solely on staging changes and safely removing the worktree itself.

What happens if pushing recovered subagent work to origin fails?

If pushing recovered subagent work to origin fails, a STATE_HANDOFF block is emitted for downstream planning. The system leaves the decision to retry or defer the push operation to the orchestrator.