AI answer monitoring · boldcreate

Headline

No engine results this run. The openrouter MCP server was not connected, so I could not query ChatGPT, Gemini, Perplexity or Claude. Google AI Overviews and Microsoft Copilot have no API, and this run was unattended. All 42 rows (7 prompts × 6 engines) are not checked, and there are no mention, citation or accuracy numbers for this period. The latest data is still the 2026-11-09 baseline run (20261109-2041_ai-answer-monitoring_boldcreate_de08): 12 of 28 mentioned, 11 of 28 cited, 5 wrong facts. That run's open issue is unverified SOC 2 claims on ChatGPT and Gemini P7.

Score

No score: the checklist does not define one, so overall_score is null.

Top 3 actions

1. Rerun this month with the openrouter MCP server connected (see _reference/mcp-and-tools.md). Use the same prompts listed in raw/engines.json, so the results compare with the 2026-11-09 baseline.
2. Confirm that the baseline's SOC 2 escalation happened (P7-chatgpt, P7-gemini) with the BoldCreate product marketing lead and the COE Team. This run could not tell whether the wrong claim is still live.
3. Check Google AI Overviews and Microsoft Copilot by hand for P1–P7 this week. Use a clean session, record the country, and fill the 14 rows. Those engines have never been checked for BoldCreate.

Findings

AI answer monitoring: boldcreate, 2026-10

Run by: Claude | Location: not set (no engine queried) | Date: 2026-11-12

Per-row entries (P1-chatgpt … P7-claude, 42 in all) are in findings.json and raw/results.csv.

Summary

• Mention rate: not checked (baseline 2026-11-09: 12 of 28; no delta possible)
• Citation rate: not checked (baseline 2026-11-09: 11 of 28; no delta possible)
• Wrong or outdated facts: not checked. The baseline found 5, and their current state is unknown:
  • ChatGPT P7 and Gemini P7: SOC 2 claims.
  • Gemini P5: trial without a credit card.
  • Perplexity P3: "creator-oriented" product.
  • Perplexity P5: retired license-holder free offer.

Gaps and fixes

This run found no new gaps. The gap list from the 2026-11-09 baseline still stands until a rerun confirms or clears it. See that run's summary for the pages and fixes.

Hand-off: there are no new deltas to post for the weekly GTM review. Report "BoldCreate AI monitoring: rerun pending". Add this run's status to the team's shared results log (location TBD).

Not checked

• ChatGPT, Gemini, Perplexity and Claude, P1–P7 (28 rows): no MCP servers were connected for this run, so list_engines and ask_engine were unavailable. I did not answer the Claude rows myself, because this agent has repo context and is not a clean session.
• Google AI Overviews and Microsoft Copilot, P1–P7 (14 rows): these have no API, and the run was unattended. They need a check by hand.
• Month-over-month comparison: there are no current values, so previous and delta_pct are null. The 2026-11-09 baseline covered roughly the same window (2026-10-12 to 2026-11-08). It is not the previous period (2026-09-17 to 2026-10-14), so I did not use it as previous.
• Accuracy checks against boldcreate.ai: I skipped them, because there were no answers to judge.
• Phase 2 hand-off and posting deltas: there was nothing new to hand off.

Proposed changes (need approval)

None from this run. The nine proposed changes from the 2026-11-09 baseline are still pending review. They include the BoldCreate trust page, a redirect or noindex for /blogs/syncfusion-license, a pricing FAQ on the card requirement, and comparison and alternatives pages.

Data sources

• Engines: none. No MCP servers were connected, the API cost was $0, and raw/engines.json records why.

• Prompt set: I reused it unchanged from the 2026-11-09 baseline. The category is "AI coding agents for engineering teams", because the checklist's BoldCreate category is still "TODO: verify category from the public site". The top competitor is Devin (Cognition), from products/boldcreate/gtm-plan.md. P6 is GTM-plan prompt 4 (WebForms to Blazor). Please confirm the category in the checklist.

• Repo references: demand-gen/search/ai-answer-monitoring.md, products/boldcreate/gtm-plan.md, and reports/publish/boldcreate/ai-answer-monitoring/20261109-2041_ai-answer-monitoring_boldcreate_de08/ (baseline context only; I copied no numbers into this run's metrics).

• raw/:

  • engines.json: the engines planned, why each was unavailable, and the exact prompts.
  • results.csv: all 42 rows, marked not checked.

  No personal data or credentials are stored.