Name: spectra-terminal
Author: ztbaker

Imported from ztbaker/spectra-terminal (AGENTS.md). Install upstream with npx skills add ztbaker/spectra-terminal. Copyright stays with the author.

AGENTS.md — SpectraTerminal

Vault Context: /Users/zacbaker/Documents/Obsidian Vault/Projects/SpectraTerminal/CLAUDE.md

Project Memory: /Users/zacbaker/Documents/Obsidian Vault/memory/project_spectra_terminal.md

Session Notes: /Users/zacbaker/Documents/Obsidian Vault/Projects/SpectraTerminal/sessions/

Main Vault MEMORY: /Users/zacbaker/Documents/Obsidian Vault/MEMORY.md

Always read these files first when starting work on SpectraTerminal. Log decisions and session notes to the vault.

MACRO Intelligence Subsystem

Overview

Macro options intelligence dashboard (MACRO command). Factor-based scoring engine that evaluates 7 instruments across 4 macro factors at 3 time horizons.

Tech Stack Additions

APScheduler (cron job, 5am ET daily)
Two-stage LLM narrative (Ollama / Claude fallback)

New Database Tables

All prefixed macro_: iv30_history, scores, regime_history, catalysts, trade_ideas, narratives, cftc_positions, input_cache

New Modules

backend/macro/data/ — Data clients (FRED, CFTC, yfinance, catalysts)
backend/macro/engine/ — Scoring, regime classification, catalyst alignment, trade filter
backend/macro/narrative/ — Two-stage LLM synthesis
backend/macro/scheduler.py — APScheduler cron
backend/macro/router.py — FastAPI endpoints
frontend/src/components/screens/macro/ — Dashboard widgets

API Endpoints

GET /api/macro/dashboard — full state
GET /api/macro/regime — regime only
GET /api/macro/catalysts — forward calendar
GET /api/macro/ideas — trade ideas
GET /api/macro/narrative — morning read
POST /api/macro/refresh — manual pipeline trigger

Cron Schedule

Daily: 5:00 AM ET
FOMC/CPI days: 5:00 AM + 6:00 AM ET

LLM Usage

Stage 1: Regime characterization (no trade suggestions)
Stage 2: Directional bias + trade commentary
Provider: Ollama (MACRO_LLM_MODEL env var) → Claude API fallback

Install spectra-terminal and it becomes one more candidate the router can pick - when it fits.

A listing is a routing target with a manifest: what it does, which domains it covers, what it costs and who publishes it. Once installed it sits beside your own models and tools, is scored like any other candidate for each request, and shows up in the trace when it wins. Ratings come from workspaces that installed it, one per account.

{ "ocm": "1", "id": "ztbaker-spectra-terminal-spectra-terminal", "kind": "skill", "name": "spectra-terminal", "description": "Always read these files first when starting work on SpectraTerminal. Log decisions and session notes to the vault.", "publisher": "ztbaker", "version": "1.0.0", "capabilities": { "domains": [ "general" ], "tags": [ "skill-md", "github-agents" ], "languages": [ "en" ] }, "quality_prior": 0.6, "examples": [ "Always read these files first when starting work on SpectraTerminal. Log decisions and session notes to the vault." ], "primary": false, "metadata": { "source": { "provider": "github-agents", "repository": "https://github.com/ztbaker/spectra-terminal", "path": "AGENTS.md", "ref": "1132dba220d82fe1e4a7538c1353cba51a1500bf", "url": "https://github.com/ztbaker/spectra-terminal/blob/1132dba220d82fe1e4a7538c1353cba51a1500bf/AGENTS.md", "key": "ztbaker/spectra-terminal/AGENTS.md" } }, "instructions": "# AGENTS.md — SpectraTerminal\n\n**Vault Context:** `/Users/zacbaker/Documents/Obsidian Vault/Projects/SpectraTerminal/CLAUDE.md`\n\n**Project Memory:** `/Users/zacbaker/Documents/Obsidian Vault/memory/project_spectra_terminal.md`\n\n**Session Notes:** `/Users/zacbaker/Documents/Obsidian Vault/Projects/SpectraTerminal/sessions/`\n\n**Main Vault MEMORY:** `/Users/zacbaker/Documents/Obsidian Vault/MEMORY.md`\n\nAlways read these files first when starting work on SpectraTerminal. Log decisions and session notes to the vault.\n\n---\n\n## MACRO Intelligence Subsystem\n\n### Overview\nMacro options intelligence das", "cost": { "context_tokens": 482 } }

spectra-terminal

About