# Strategy Reset

> Use when three same-strategy attempts fail, two approaches oscillate without evidence, or the current abstraction and evidence source cannot resolve the blocker.

- Skill: `ihabkhaled/strategy-reset` (Agent Skill)
- Install (CLI): `npx skillmds@latest add ihabkhaled/strategy-reset`
- Raw SKILL.md: https://api.skillmd.com/api/skills/ihabkhaled/strategy-reset/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: ihabkhaled (https://skillmd.com/u/ihabkhaled)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/ihabkhaled/strategy-reset

---


# Strategy Reset

## Core principle

A reset must introduce real novelty: a new hypothesis, source, isolation, layer, or execution strategy.

## Procedure

1. Freeze the failed and oscillating approaches.
2. Summarize evidence and invalidated assumptions.
3. Name what all attempts had in common.
4. Select a materially different strategy using objective criteria.
5. Reduce the problem to the smallest discriminating test.
6. Execute once and update the parent task.

## Required output

Failed strategy, information gained, new strategy, discriminating action, and result.

## Limits

One reset action before reassessment; no fourth equivalent retry.

## Common mistakes

Do not change only command syntax, tool brand, or wording while preserving the same hypothesis.

## Stop condition

Stop when the new strategy produces evidence or proves a real blocker.