GlassBots Diagnostic — Example Logistics

Generated 2026-08-20 · Tier 4 (workspace-folder analysis) · 20 findings · raw generated output (.md)

37 / 100
FAILING

Recommendation: Rebuild. Salvage the knowledge, not the machinery.

Handholding Ratio: not measurable from a capture alone — requires the interview tier to convert stalls and interventions into hours per completed item. (Reported basis: structural evidence only.)

The three biggest causes

1 · Authority system answers "may I?" — nothing answers "what is done?" Deny-by-default rules with tamper-evident hashes exist; nothing defines a finished task. Stopping early violates no rule — so the safe behaviour is to stall, and a person pushes everything through.
2 · No machine-checkable definition of "finished" anywhere No acceptance criteria, no test gates, no named-output contracts. "Done" is decided by a person looking at things — so the system can stop early and nothing detects it.
3 · Scheduled jobs failing repeatedly Work the system scheduled for itself is erroring on schedule, not completing — and nothing escalates it.

Scores — the ten checks

Completion0/10
Handholding5/10
Idle Burn5/10
Approval Drag5/10
Definition of Done2/10
Invented Work7/10
Visibility6/10
Editability7/10
Resilience0/10
Lock-in & Exposure0/10

Where the work stops

TODAY — silent stall task starts partial work stops quietly no alarm · no owner found weeks later by accident WITH GLASSBOTS — bounded loop task starts acceptance check fail → exact feedback retry ≤ 3× done, or a human told exactly what's missing

The same task under both regimes. The difference is not intelligence — it is a checkable definition of finished.

Selected findings (of 20)

SeverityFinding
CRITICALUnattributed service listening on all interfaces
CRITICALNo offsite backup tooling evidenced anywhere in the capture
CRITICALAcceptance criteria verify shape, not substance — owners must supply the rest
HIGHScheduled jobs failing repeatedly in the last 48 hours
HIGHFive persistent agent personas with independent state
HIGHSingle alerting and control channel
HIGHSource projects not under version control
HIGHArchitecture document describes design but no operational metrics

Every finding is evidenced

The full report traces each finding to a specific file in the captured system, marks everything Verified (measured) vs Reported (described), and states plainly what this access tier could not assess. Reproducible: the same capture re-scored yields the same report. Read the raw generated output →
Book your Diagnostic — US$199 / AU$275

Start free: the 2-minute scorecard

Sample generated by GlassBots on a synthetic system — your report is produced by the same tool against your captured workspace.