lots of change - all to start my 3rd redo

worker-toolkit-potion-polyglot-git-archeology/explore/plugins/create-snapshot/skills/snapshot/SKILL.md

@@ -0,0 +1,76 @@
+---
+name: snapshot
+description: Capture the current conversation and repo state as a snapshot, to be replayed as a task. Use when the user wants to snapshot a mistake the agent just made.
+---
+
+# Create Snapshot
+
+You are capturing a snapshot of the current conversation and repo state so it can be
+replayed as an RL training task.
+
+## Step 0: Mark where the snapshot begins
+
+Run this FIRST, before asking anything. It records where the conversation ended so the
+questions below aren't captured as part of it:
+
+```bash
+"${RACCOON_SNAPSHOT_PLUGIN_ROOT:-/workspace/plugins/create-snapshot}/bin/capture-snapshot.mjs" \
+  --mark-start --harness "${RACCOON_HARNESS:?not set — start your session through the launcher (the plain agent command, e.g. \`codex\`) so the snapshot records which agent it came from}"
+```
+
+## Step 1: Ask annotation questions
+
+**Important — tell the user this first, verbatim:**
+
+> ⚠️ This snapshot captures your entire conversation history with me, not just the most
+> recent turn. If you told me the answer earlier in this conversation, or steered me
+> toward it, the agent will see that same context when the snapshot replays — and will
+> probably solve the task without making the mistake. Your task will be contaminated.
+>
+> If you've leaked the answer at any point in this conversation: if your agent can rewind
+> (Claude Code's `/rewind`), rewind to a point before the contamination and snapshot from
+> there. If it can't — codex has no rewind — this snapshot is not salvageable: start a
+> fresh session, reproduce the mistake without steering, and snapshot that instead.
+
+Wait for the user to acknowledge before moving on.
+
+Ask the user each of these questions **one at a time** as plain text, waiting for their
+response before proceeding to the next:
+
+1. "What were you trying to do?"
+2. "What were you hoping was going to happen?"
+3. "What did the agent actually do instead?"
+
+## Step 2: Propose a slug
+
+Based on the user's answers, generate a **short kebab-case slug** (2-4 words) that
+captures the essence of the mistake. For example: `bad-refactor`,
+`wrong-test-strategy`, `missed-edge-case`.
+
+Present your suggestion and ask the user to confirm or provide an alternative.
+
+## Step 3: Write annotation file and run capture
+
+Write the annotation to a temporary JSON file, then run the capture script.
+
+Write this JSON to a temp file (use a path like `/tmp/snapshot-annotation-<timestamp>.json`):
+
+```json
+{
+  "what_trying": "<answer to question 1>",
+  "what_hoping": "<answer to question 2>",
+  "what_happened": "<answer to question 3>"
+}
+```
+
+Then run:
+
+```bash
+"${RACCOON_SNAPSHOT_PLUGIN_ROOT:-/workspace/plugins/create-snapshot}/bin/capture-snapshot.mjs" \
+  --harness "$RACCOON_HARNESS" \
+  --slug <slug> \
+  --annotation <temp-file-path> \
+  --output-dir /workspace/snapshots
+```
+
+Report the script's stdout output verbatim to the user. Do not paraphrase or shorten paths.