Ship the batch: undo, parallel subagents, lazy MCP, hot-reloaded skills; docs and CI/CD

Loop and ergonomics batch across Now/Next and Maintenance:
- /undo and /redo via pre-prompt file snapshots (snapshot.ts)
- task takes a tasks[] array and runs investigations concurrently (subagent.ts)
- lazy MCP tools: mcp_list/mcp_inspect/mcp_call meta-tools, eager opt-in (mcp.ts, config.ts)
- skill tool reads its list live so a mid-session install is callable next turn (skills.ts)
- tool-name lists (tool-kinds.ts) derived from a mutating() marker; gates previously ungated writes
- prune/session recovery path summarized, and step-back doom-loop primitive (step-back.ts)
- @file completion re-walks on a slow cooldown; estimateTokens and pricing labeled as estimates

Docs: README, CHANGELOG, docs/{mcp,architecture,development} updated to match.
CI/CD: bun install-store cache and concurrency gates on both workflows; release.yml now
composes file-based release notes via scripts/make-release-notes.ts and verifies every binary.

test/prompt.test.ts

@@ -33,6 +33,13 @@ test('mcp tools are grouped with their naming convention explained', () => {
   expect(rendered).toContain('needs approval');
 });
 
+test('lazy mcp meta-tools prompt the workflow, and mcp__ and meta-tools do not double-list', () => {
+  const rendered = renderTools(['read_file', 'mcp_list', 'mcp_inspect', 'mcp_call']);
+  expect(rendered).toContain('mcp_list, mcp_inspect, mcp_call');
+  expect(rendered).toContain('Never guess a server or tool name');
+  expect(rendered).not.toContain('mcp__<server>__<tool>');
+});
+
 test('an unknown tool is listed rather than silently dropped', () => {
   expect(renderTools(['read_file', 'some_plugin_tool'])).toContain('some_plugin_tool');
 });