How safe the fleet is — the score, what lowers it, and where it heads.

The board read on agent safety: one score, the factors dragging it down, and the agents to harden first.

Bottom line

Fleet posture 76/100 (target 80). Biggest drag: No tool-approval policy (−10 each, 5 agents). Worst agent: marketing-gpt (25). Hardening those moves the score most.

Do nextHarden worst agent — marketing-gpt →Fix "No tool-approval policy" on 5 agents

fleet posture score

target ≥ 80

critical agents

score < 50

88%

instrumented coverage

7/8 agents

shadow installs

installed, unmonitored

Fleet posture score

instrumented agents · target ≥ 80

76/ 100

0target ≥ 80100

Hardened

At risk

Critical

What lowers the score

risk factors across the fleet

No tool-approval policy

5−10 each

Running as root

2−25 each

Approvals & sandbox bypassed (yolo)

1−35 each

Sandbox disabled

1−20 each

Permission checks bypassed

1−15 each

Posture regression — current vs worst observed

agents that have been worse than they are now

support-copilotworst observed 50 (Running as root)current 65↑ 15

Critical & at-risk agents

score < 80 · lowest first

marketing-gptApprovals & sandbox bypassed (yolo) · Running as root · Permission checks bypassed25 Zed AgentNo tool-approval policyshadow60 support-copilotRunning as root · No tool-approval policy65 Codex CLISandbox disabled · No tool-approval policy70

At-rest installations — coverage

4/5 instrumented · 1 unmonitored

AdapterVersionExecutableConfigCoverage

claude-cli2.1.161/Users/mark/.local/bin/claudepresentinstrumented

claude-desktop1.9255.2/Applications/Claude.app/Contents/MacOS/Claudepresentinstrumented

cursor-cli3.5.17/Applications/Cursor.app/Contents/MacOS/Cursorpresentinstrumented

cursor-desktop3.5.17/Applications/Cursor.app/Contents/MacOS/Cursorpresentinstrumented

zed0.229.0/Applications/Zed.app/Contents/MacOS/zedpresentunmonitored

Fields: security_score · factors · worst_score · worst_factors · danger_level · instrumented · installations — from app/data/c16-fleet.ts (qcontrol /api shape). Score = 100 − sum of FACTOR_PTS deductions.