OpenSkillIndex← back to search
claude-codeautomation

Claude-skill-registry dynamic-budget-orchestrator

Dynamically scale model token budgets using resource telemetry, prompt size, and profile presets. Use when token limits must adapt to hardware constraints, per-request size, or safe/fast/quality modes.

install
source · Clone the upstream repo
git clone https://github.com/majiayu000/claude-skill-registry
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/majiayu000/claude-skill-registry "$T" && mkdir -p ~/.claude/skills && cp -r "$T/skills/data/dynamic-budget-orchestrator" ~/.claude/skills/majiayu000-claude-skill-registry-dynamic-budget-orchestrator && rm -rf "$T"
manifest: skills/data/dynamic-budget-orchestrator/SKILL.md
tags
#token-budgeting#resource-management#optimization#telemetry#automation#llm-scaling
source content

Dynamic Budget Orchestrator

Use this skill to make LLM token limits elastic to local resources and request size.

Workflow

  1. Capture RAM and VRAM snapshot (best effort).
  2. Estimate prompt size and compute a prompt-based cap.
  3. Apply a profile factor (safe, fast, balanced, quality).
  4. Clamp to min/max and emit an effective token budget.

Scripts

  • Run: python skills/automation/dynamic-budget-orchestrator/scripts/probe_budget.py

References

  • references/profile_presets.json