⚠️ SUPERSEDED (2026-07-04, banner added 2026-07-06). The evidence and grades in this dossier remain valid and citable. The recommendation (fork simple10; graft cast
analytics.ts) is superseded: the project is a greenfield clean build per../analysis/best-path-decision.md, and cast/disler/nirdiamant code is clean-room-only per CD-9. Current entry point:../analysis/PROJECT-STATE-2026-07-06.md.
What this is: an independent, source-level audit of the six candidate dashboards
evaluated for agenthropic, plus a meta-audit of the two vendor “Technical
Due-Diligence” reports (v1, v2.0 extended) that recommended one of them.
Bottom line: the vendor panel’s own scoring model ranks simple10 first, then
overrides itself to recommend hoangsonww on a tie-break that rests on a factual
error. Corrected, the recommendation flips. My independent grades diverge
systematically from the panel: it over-rated the flashy-but-fragile and under-rated
the disciplined.
Status: analysis only. No adopt/fork/build action has been taken. The direction decision waits on Ivan.
| File | What’s in it |
|---|---|
| methodology.md | How the audit was run, sources, tools, limits, the verified GitHub fact base |
| report-meta-audit.md | The panel’s self-contradiction, the fact-check table, the weighted scoring model dissected |
| market-landscape.md | Selection bias — the popular tools the panel never mentioned — and the real market gap |
| security.md | Cross-cutting security reality (0.0.0.0, no-op auth, RCE, SSRF) + Mac Mini hardening checklist |
| recommendation.md | The decision, its rationale, the step plan, alternatives, and open questions |
| projects/simple10.md | A− — my pick as base |
| projects/hoangsonww.md | B− — study, don’t adopt blind |
| projects/cast.md | C — two ideas worth stealing, else lock-in |
| projects/disler.md | C− — teaching example, no tests, no license |
| projects/nirdiamant.md | C+ — wrong shape (flat feed) |
| projects/claude-code-templates.md | The 28.4k★ tool the panel missed — the baseline to differentiate against |
A one-page executive synthesis also lives at ../independent-due-diligence.md.
| Project (owner) | Panel v1 | Panel v2 (/5) | Independent | Move |
|---|---|---|---|---|
| simple10/agents-observe | B+ | 4.1 (highest) | A− | ▲ |
| hoangsonww/Claude-Code-Agent-Monitor | A− | 4.0 | B− | ▼ |
| ek33450505/claude-code-dashboard (CAST) | B− | 3.8 | C | ▼ |
| disler/…observability | B+ | 2.7 | C− | ▼▼ |
| NirDiamant/claude-watch | C+ | 2.6 | C+ | = |
| davila7/claude-code-templates (missed) | — | — | C (for this need) | new |
Fork simple10, harden it (127.0.0.1 + token + launchd, no Docker), then graft
hoangsonww’s Telegram webhook provider and cast’s delegation-savings metric —
with hoangsonww kept cloned only as a reference for its richer D3 graph. Full
rationale in recommendation.md.
All figures verified at source on 2026-07-03. These repos move fast — re-verify before any final commitment.