5 items
The Eject Button for ChatGPT Doom Loops - Rescue poisoned conversations 🚑 Stop wasting hours when the AI keeps apologizing, repeating broken code, or going in circles. Context Ambulance scrapes your "poisoned" conversation, identifies what went wrong, and generates a clean rescue package you can paste into a fresh chat (or switch to Claude / Gemini) to continue productively. 🔥 THE PROBLEM You know this nightmare: • Hour 3: "Let me try a different approach…" (again) • Hour 4: More apologies + the same broken fix • Hour 5: You don't want to restart, but you're stuck LLMs can become "context-poisoned" after repeated failed attempts. Starting fresh usually means losing progress. Until now. ✨ HOW IT WORKS (3 CLICKS) Click the ambulance icon while on ChatGPT / Claude / Gemini Auto-scrapes the conversation (no copy/paste) Choose your action: 🚑 Rescue — Download a clean .md package for a fresh LLM session 🔄 Revival — Copy a reset prompt to paste in the same chat 💾 Export — Save the full conversation as a clean .md file Rescue packages include: – Original goal (extracted) – Primary blocker (root cause) – Cleaned history (loops removed) – Next-step recommendations Then paste the rescue package into a new chat. Same context, none of the poison. 💾 NEW: EXPORT CONVERSATIONS Switching LLM providers? Want to keep a conversation for your records? Export any chat from ChatGPT, Claude, or Gemini as a clean markdown file — no analysis, just a straight transcript. Files are named with the conversation title automatically. 🎯 WHAT GETS RESCUED ✅ Apology cascades ✅ Code churn / back-and-forth edits ✅ Repetitive errors ✅ Dead-end tangents ✅ Partial solutions buried in noise 🔒 PRIVACY FIRST • No tracking, no account required • Conversations are not stored on our servers • Open source — audit the code on GitHub 🧰 OPTIONAL: PYTHON CLI Power users can install the companion CLI for offline analysis, custom rules, and batch processing. Open-source repo: github.com/leyiwu/context-ambulance Chrome Web Store listing: (this page)
Mar 10, 2026
rating_count is the Chrome Web Store ratings count, not a written-review count.
Media assets
Screenshots and videos on the listing.
Has promo video
Whether the listing includes at least one video.
Languages
Declared language locales.
Developer website
Listing exposes a developer website URL.
Contact email
Listing exposes a contact email.
Keyword in name
Case-insensitive substring match in the name.
Keyword in description
Case-insensitive substring match in the description.
Keyword occurrences in description
Count of case-insensitive occurrences in the description.
Category user-count percentile
Share of same-category extensions with fewer users (null if unknown).
These are transparent listing completeness / keyword signals, not a prediction of Chrome Web Store search ranking.