Menu
Router for the five-tier fleet starter skill (l1-basic..l5-autonomous) — selects the AAL/complexity level and its context budget for a spawned agent.
Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.
Basé sur la classification professionnelle SOC
Question Suppression Ladder + L1 Inline Resolution Rule + Hard-gated Action Boundary. Activated by DATARIM_AUTO_MODE=1 + a per-task autonomous-mode marker.
Use when implementation is complete and all tests pass. Decides how to integrate the work via merge, PR, or cleanup with structured options.
Canonical S1–S9 security rule reference cited from CLAUDE.md § Security Mandate. Load for plan/qa/compliance/do touching shipped artefacts.
Technical rules for publishing content to social media and websites — platform limits, formatting, API patterns, OG tags, multi-platform workflow.
Testing pyramid, frameworks, mocking. Load first; then the fragment for the active gate (live smoke, silent failure, bats, legacy triage).
Operator wishlist checklist seeded at /dr-prd or /dr-plan; verified at /dr-qa and /dr-compliance with BLOCKED routing on missed/partial without override.
| name | fleet |
| description | Router for the five-tier fleet starter skill (l1-basic..l5-autonomous) — selects the AAL/complexity level and its context budget for a spawned agent. |
| model | inherit |
| current_aal | 1 |
| target_aal | 2 |
Fleet workers receive a task brief plus one complexity-tier starter skill that sets their context budget and operating rules. This entry is a router: pick the single level fragment that matches the task's complexity, load it, and follow it. Do not load more than one level.
| Level | When | Context budget | Fragment |
|---|---|---|---|
| L1 — basic | Single elementary step, one tool, no branching | ~200 tokens | l1-basic/SKILL.md |
| L2 — structured | A few templated steps, allowed-command list | ~500 tokens | l2-structured/SKILL.md |
| L3 — analyst | Analysis with a choice between alternatives; RAG on demand | ~1500 tokens | l3-analyst/SKILL.md |
| L4 — expert | Complex task with several subtasks; decompose and sequence | ~4000 tokens | l4-expert/SKILL.md |
| L5 — autonomous | Self-managed multi-subtask coordination; KB pre-fetch | ~6000 tokens | l5-autonomous/SKILL.md |
level-mismatch: needs >=Lx
instead of improvising above its tier.forbidden_actions or any hard-gated action boundary.