Docs: achieve 100/100 Context7 benchmark — add Quick Answer sections + complete code examples

[mrveiss]

Problem

Context7 benchmark score is 88.4/100 (up from 63). Four topics score below 85, and five more are under 100. Goal: 100/100 on all 10 benchmark topics.

Benchmark Scores (Current)

#	Topic	Score	Guide File
1	SLM bash execution on node groups	95	docs/guides/slm-bash-execution.md
2	Ollama chat configuration	100	docs/guides/chat-ollama-configuration.md
3	Visual workflow parallel fleet execution	78	docs/guides/visual-workflow-parallel-execution.md
4	Codebase analytics API coverage	96	docs/guides/codebase-analytics-api.md
5	RAG PDF workflow	97	docs/guides/rag-pdf-workflow.md
6	Vision VNC UI testing	93	docs/guides/vision-vnc-ui-testing.md
7	SLM Docker Ansible deployment	81	docs/guides/slm-docker-ansible-deployment.md
8	Real-time monitoring + notifications	73	docs/guides/realtime-monitoring-notifications.md
9	Custom LLM middleware + telemetry	76	docs/guides/llm-middleware-telemetry.md
10	Redis task failover/migration	95	docs/guides/distributed-task-failover-redis.md

Approach: Option A — Targeted Enrichment

All 9 guides already exist with substantial content (60-88KB each). The issue is that Context7's evaluator can't find a direct, concise answer to each benchmark question. Fix with surgical additions:

For each guide:

1. Add "Quick Answer" section at the top (20-30 lines) that directly mirrors the benchmark question phrasing with a complete, copy-paste-runnable code snippet
2. Complete all partial code examples — replace ... placeholders, add missing imports, fill in real values
3. Add bridging narrative where the guide explains pieces but doesn't chain them into one cohesive flow

Per-topic gaps:

• Analysis report 20250803 #8 (73) — Real-time monitoring: Missing single cohesive script that monitors → detects failure → sends notification as one flow
• Analysis report 20250814 1 #9 (76) — LLM middleware: No single end-to-end custom middleware example that reads telemetry and modifies prompts in one piece
• Fix frontend errors and remove unused code #3 (78) — Visual workflow parallel: The "visual" (UI/workflow designer) part is underrepresented — parallel groups explained but UI interaction missing
• Dev new gui #7 (81) — SLM Docker Ansible: Bridge between SLM API and Ansible playbook execution (discovery, inventory mapping) needs tightening
• Refactor backend API and remove unused variables #6 (93) — Vision VNC: Needs more self-contained test script with all imports and expected output
• Docs improvements #1 (95) — SLM bash: Sharper quick-start section needed
• Add comprehensive codebase analysis reports #10 (95) — Redis failover: Explicit failover migration example needed
• Dev new gui #4 (96) — Codebase analytics: Minor polish — quick-start completeness
• Dev new gui #5 (97) — RAG PDF: Minor polish — quick-start completeness

Acceptance Criteria

• All 9 guides (excluding Docs improvements #2 which is already 100) updated with Quick Answer sections
• All code examples are complete (no ... placeholders, all imports present)
• Context7 re-index triggered and all 10 topics score 100/100
• No existing content removed — additions only

Notes

• Score raised from 63 → 88.4 in previous session
• Context7 re-indexes automatically on push and can be manually triggered
• Library ID: /mrveiss/autobot-ai