Bug Report
Summary
CLEO has two orthogonal urgency axes that operators conflate:
priority ∈ {critical, high, medium, low}
severity ∈ {P0, P1, P2, P3} — triggers Ed25519 attestation
Operator instructions colloquially use "P0" for the most urgent. CLEO's priority enum doesn't use P-codes; CLEO's severity does. There's no single-axis way to ask "show me everything truly urgent" — you must check both. Worse: --severity P0 silently downgrades without a signing key (separate bug).
Reproduce
cleo show T149 | jq '.data.task | {priority, severity}'
# → {"priority":"critical","severity":"P1"}
# No "show me urgent" command
cleo find --urgent # ← does not exist
cleo next | jq '.data.nextTasks[].score'
# → priority is in the score, severity is not weighted
# cleo briefing surfaces tasks by `nextSuggested` heuristic — severity P0/P1 not boosted
cleo briefing | jq '.data.nextTasks | map(.id)'
Expected
cleo find --urgent returns the union of priority IN (critical, high) OR severity IN (P0, P1).
cleo briefing highlights tasks with priority=critical OR severity ∈ {P0,P1} in a dedicated ⚠ urgent section.
cleo update --help documents the dual-axis side-by-side with examples of when to use which.
cleo show <id> renders urgency line: Urgency: priority=critical · severity=P1 · deadline=2026-06-01.
Actual
The two axes are documented as "orthogonal" but operators have no unified read surface. Result: T148/T149 (P0-deadline tasks) currently show as priority:critical, severity:P1 and cleo next returns generic "score:50" tasks ahead of them.
Impact
Urgent work gets buried. Resuming agents check cleo next, see no P0/P1 in the top suggestions, and start on whatever the score-50 default surfaces. Time-critical bugs (T148 Gemini sunset 2026-06-01) need explicit human flagging to be picked up.
Suggested fix
- Add
--urgent flag to cleo find.
- Boost
severity ∈ {P0,P1} in cleo next scoring (additive +30/+15).
- Section-header surface in
cleo briefing.
- Document axes in
cleo update --help:
--priority Workload urgency (critical|high|medium|low) — orchestrator scheduling weight
--severity Bug/risk classification (P0|P1|P2|P3) — triggers signed attestation
Cross-ref
.cleo/research/sg-sor/06-cleo-gaps.md issue #9.
Severity: Moderate
Area: cli
Are you using an AI agent?
Yes - AI agent filed this issue
Environment
| Component |
Version |
| CLEO |
2026.2.1 |
| Node.js |
v24.13.1 |
| OS |
linux 6.19.14-200.fc43.x86_64 x64 (x64) |
| Shell |
/bin/bash |
| gh CLI |
gh version 2.87.3 (2026-02-23) |
| Install |
/home/keatonhoskins/.npm-global/bin/cleo |
Bug Report
Summary
CLEO has two orthogonal urgency axes that operators conflate:
priority∈ {critical, high, medium, low}severity∈ {P0, P1, P2, P3} — triggers Ed25519 attestationOperator instructions colloquially use "P0" for the most urgent. CLEO's
priorityenum doesn't use P-codes; CLEO'sseveritydoes. There's no single-axis way to ask "show me everything truly urgent" — you must check both. Worse:--severity P0silently downgrades without a signing key (separate bug).Reproduce
Expected
cleo find --urgentreturns the union ofpriority IN (critical, high)ORseverity IN (P0, P1).cleo briefinghighlights tasks withpriority=criticalORseverity ∈ {P0,P1}in a dedicated⚠ urgentsection.cleo update --helpdocuments the dual-axis side-by-side with examples of when to use which.cleo show <id>renders urgency line:Urgency: priority=critical · severity=P1 · deadline=2026-06-01.Actual
The two axes are documented as "orthogonal" but operators have no unified read surface. Result: T148/T149 (P0-deadline tasks) currently show as
priority:critical, severity:P1andcleo nextreturns generic "score:50" tasks ahead of them.Impact
Urgent work gets buried. Resuming agents check
cleo next, see no P0/P1 in the top suggestions, and start on whatever the score-50 default surfaces. Time-critical bugs (T148Gemini sunset 2026-06-01) need explicit human flagging to be picked up.Suggested fix
--urgentflag tocleo find.severity ∈ {P0,P1}incleo nextscoring (additive +30/+15).cleo briefing.cleo update --help:Cross-ref
.cleo/research/sg-sor/06-cleo-gaps.mdissue #9.Severity: Moderate
Area: cli
Are you using an AI agent?
Yes - AI agent filed this issue
Environment