# Temporal

> Orchestrates temporal CV sub-skills (LPC, feature-point tracking) for Viagent. A vision LLM uses preview frames plus this file's viagent_routing to pick which sub-skills run.

- Skill: `iatomicreactori/temporal` (Agent Skill, multi-file: 4 files)
- Install (CLI): `npx skillmds@latest add iatomicreactori/temporal`
- Raw SKILL.md: https://api.skillmd.com/api/skills/iatomicreactori/temporal/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- License: Proprietary - Viagent
- Author: iatomicreactori (https://skillmd.com/u/iatomicreactori)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/iatomicreactori/temporal

---


# Temporal analyst - skill orchestration

Executable code: `src/skill/temporal/`. This file defines **when** each sub-skill is appropriate for LLM or explicit routing.

```yaml
viagent_routing:
  version: 1
  subskills:
    - id: lpc
      when: >-
        Local phase coherence around detected faces across frames. Use when faces
        or vlogs/talking-head are plausible. Omit for scenery-only, slides, or
        screen recordings with no people.
    - id: feature_stability
      when: >-
        Feature point tracking stability across frames. Useful when there is
        texture and motion; weak for static slides or blank backgrounds-may still
        run if motion exists.
```

## Runtime

Sub-skills are chosen by the **vision router** (preview frames + this file). Optional tuning: `llm.skill_router_temperature`, `llm.skill_router_max_preview_frames`.