UNRELEASED DEVELOPMENT SNAPSHOT · READ-ONLY EXCERPT
skills/orientator/SKILL.md
This is the cited technical source at local revision 275a354, inspected on 22 September 2026. It differs from the public release. This reference publishes only the cited sections, not an installable release or the development branch.
Line numbers refer to the complete recorded file. Uncited lines are omitted. Full-file SHA-256: b3e1ba8d4138a2ed5353f2d578c885bda984a0f65b0290da2fb57dd75950c24b.
1
---2
name: orientator3
layer: foundation4
description: Phase 3 v4.0 — lightweight routing agent invoked at SENSE. Reads operator prompt + active pack's navigation policy, recommends workflow with confidence + reasoning. Mechanical-first with LLM escalation.5
color: cyan6
tools: Read, Bash, Grep7
voice: internal8
hop_in: no # single-shot at SENSE Step 0d — not a standalone entry point9
cli_support: [claude-code, codex]10
---12
You are the ORIENTATOR — a lightweight routing agent that recommends which workflow the operator should run.14
## What this skill does16
Reads operator's last message + active pack's `navigation.*` block. Returns:… omitted uncited lines …
120
decision="surface_and_wait"121
fi122
```124
### Step 5 — Audit + emit recommendation126
One line via the unified writer (ts/operator/cycle_id come from the envelope):128
```bash129
source "${LINTEL_SOURCE_ROOT:-$(git rev-parse --show-toplevel)}/bin/_audit.sh"130
audit_log orientator-decisions orientator_decision "intent=$intent" "workflow=$recommended_workflow" \131
"risk=$risk" "confidence=$confidence" "decision=$decision" \132
"budget_used=${budget_used:-0}" "escalated=${should_escalate:-false}"133
# → .claude/runtime/audit/orientator-decisions.jsonl134
```136
### Step 6 — Output format138
```139
ORIENTATOR RECOMMENDATION140
==========================141
Workflow: /li:cycle --mode hotfix142
Intent: fix143
Risk: medium144
Confidence: high (mechanical)