Recommendations

Agent credential rotation has stopped on something outside the workflow, so no gate or dependency will clear it. Blocked at Build — Four standing credentials belong to a vendor whose API caps rotation at 10 calls/hour. Rotation cannot complete inside the maintenance window until that cap is lifted.. Nothing downstream will move until it clears.

Ines Marchetti (CAO) is holding 3 decisions, 2 of them rated high risk. Each one is work somebody else has already stopped doing while they wait.

Nadia Brecht (CKO) is holding 3 decisions, 1 of them rated high risk. Each one is work somebody else has already stopped doing while they wait.

Junia Falkenrath (CMO) is holding 4 decisions, 2 of them rated high risk. Each one is work somebody else has already stopped doing while they wait.

Vale Ordonez (COO) is holding 3 decisions, 1 of them rated high risk. Each one is work somebody else has already stopped doing while they wait.

Soren Aldiss (CPO) is holding 3 decisions, 1 of them rated high risk. Each one is work somebody else has already stopped doing while they wait.

Blocked employees hold their assignments without progressing them, so the work is neither moving nor available to anybody else. They span 2 departments.

The estate is carrying 19 unresolved gates, 6 of which are holding work up right now. Dai Okonjo holds 4 of them. Governance is the constraint, not capacity.

Engineering is carrying 2 objectives it has flagged itself — 1 blocked, 1 at risk. These are the department's own commitments, so the assessment is not the engine's opinion.

Research is carrying 2 objectives it has flagged itself — 1 blocked, 1 at risk. These are the department's own commitments, so the assessment is not the engine's opinion.

7 of 8 in-flight workflows are either blocked or paused at a gate. At this share the constraint is systemic — clearing one workflow will not move the estate.

AI Automation Agency has 1 projects stopped at 41% mean progress across 1 department. That is spent effort producing nothing until each one is cleared.

Ecommerce Brand has 1 projects stopped at 33% mean progress across 1 department. That is spent effort producing nothing until each one is cleared.

Media Company has 1 projects stopped at 44% mean progress across 1 department. That is spent effort producing nothing until each one is cleared.

7 of 127 agents in the fleet carry blanket grants rather than an enumerated tool list, 5 of which can write. That may be correct for each of them — but nothing currently re-checks that it still is.

Engineering has 1 projects it cannot progress, carrying 44% mean progress between them. Work that is part-done and stopped is the most expensive state work can be in — the investment is spent and the return has not arrived.

Research has 1 projects it cannot progress, carrying 62% mean progress between them. Work that is part-done and stopped is the most expensive state work can be in — the investment is spent and the return has not arrived.

Sales has 1 projects it cannot progress, carrying 29% mean progress between them. Work that is part-done and stopped is the most expensive state work can be in — the investment is spent and the return has not arrived.

CAO holds 4 objectives and has flagged 2 as at risk or blocked. At the executive layer these are enterprise commitments, not team goals — a missed one moves a number the CEO reports.

COO holds 4 objectives and has flagged 2 as at risk or blocked. At the executive layer these are enterprise commitments, not team goals — a missed one moves a number the CEO reports.

CRO holds 4 objectives and has flagged 2 as at risk or blocked. At the executive layer these are enterprise commitments, not team goals — a missed one moves a number the CEO reports.

44 of 44 memory records are not embedded, so a retrieval query cannot reach them. Every agent drawing on memory is answering from 0% of what the enterprise actually knows, and has no way to tell that it is doing so.

9 of 43 AI employees (21%) are idle or under 20% load, while Operations runs at 68%. The enterprise is paying for capacity it is not routing work to.

These employees sit at or above 80% while their departments read as having headroom. A departmental mean cannot see them, so nothing else in the app will raise this — and each one is a single point of failure for whatever they are holding.

Nadia Brecht scores 55 against a leadership mean of 73. The score is mean attainment across the seat's own metrics, so this is a statement about targets being missed rather than about the person holding the seat.

More than one installed agent answers to performance-optimizer. Workflow tasks assign agents by display name, so a task naming one of these cannot be resolved to a single installation without its plugin scope.