mcpField
A
agent-self-recovery

agent-self-recovery

github.com/archieindianAgent workflows7347 installsUpdated 26w ago
Repository

About agent-self-recovery

A recovery procedure for agents that are repeating attempts or losing sight of a task goal. It records the current state, identifies the stuck pattern, and directs the agent to try a different approach, seek help, or hand off the task.

Key features

  1. 01Detects repeated attempts and lost goals
  2. 02Records attempts and state in a file
  3. 03Classifies common stuck patterns
  4. 04Selects recovery steps by problem type
  5. 05Escalates blocked tasks for assistance

Use cases

  1. 01Repeated fixes produce the same result
  2. 02Each attempted fix reveals another dependency
  3. 03Debugging when the error is unclear
  4. 04Refocusing an agent that lost the original request

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/archieindian/openclaw-superpowers --skill agent-self-recovery

After installing, type/agent-self-recoveryin Claude Code to confirm it loaded.

Bundled filesSTATE_SCHEMA.yaml

Details

From the project README

Agent Self-Recovery

State file: ~/.openclaw/skill-state/agent-self-recovery/state.yaml

Signs You're Stuck

  • Tried the same approach 3+ times with the same result
  • Making changes that feel random rather than targeted
  • Lost track of the original goal
  • Taking longer than expected on a 'simple' step

Recovery Protocol

Step 1: Stop

Stop making changes. No more quick tries.

Step 2: Write State

Write to the state file:

  • what_trying: what were you trying to do?
  • what_tried: append this latest approach to the list
  • loop_type: name the pattern (see Step 3)
  • attempts: increment by 1
  • status: recovering
  • started_at: current timestamp (first time only)
Step 3: Name the Loop

Pick the loop_type:

  • same-fix-variations — trying the same fix with minor changes
  • cascading-dependencies — each fix reveals a new dependency
  • unclear-error — unclear what the actual error is
  • lost-goal — lost track of original requirement
Step 4: Break the Loop

If loop_type: unclear-error → use systematic-debugging first. If loop_type: lost-goal → re-read the original requirement. If genuinely blocked → document clearly, ask the user. Update state status: escalated. If context too large → use context-window-management. If truly lost → use task-handoff, ask user to redirect. Update state status: escalated.

Step 5: Resume with a New Approach

Choose a genuinely different approach. Update state last_approach, append to what_tried. On success: update state status: recovered.

Comments

Markdown supported