Mortdecai b901b8be6e fix: use fast model for story selection, gemma4 for summaries only ...

gemma4:26b under format:json ignores the bare-array instruction and returns a
reasoning object ({"thought": ...}), which broke Phase-1 selection (Unterminated
string -> silent fallback to first-5). Selection now uses ollama_select_model
(default qwen2.5:1.5b); summaries stay on gemma4:26b. Also tolerate dict-shaped
selection responses defensively.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

2026-06-07 13:36:20 -04:00

docs/superpowers

docs: implementation plan for HA/Google Home POS briefing trigger

2026-06-07 13:07:10 -04:00

scripts

docs: design for HA/Google Home on-demand POS briefing trigger

2026-06-07 13:01:15 -04:00

tests

fix: add tests/pytest.ini + move Path import to top (completes 2308f24)

2026-06-07 13:28:15 -04:00

.gitignore

Initial commit: POS-only daily briefing for Epson TM-m30

2026-03-17 00:15:49 -04:00

.secrets.baseline

feat: summarize with gemma4:26b using converged prompt

2026-06-07 13:17:28 -04:00

config.example.json

fix: use fast model for story selection, gemma4 for summaries only

2026-06-07 13:36:20 -04:00

CONTEXT.md

feat: hourly weather forecast + steel141 migration notes

2026-06-07 13:15:02 -04:00

pos_briefing.py

fix: use fast model for story selection, gemma4 for summaries only

2026-06-07 13:36:20 -04:00

SESSION.md

feat: hourly weather forecast + steel141 migration notes

2026-06-07 13:15:02 -04:00

S

Description

POS-only daily news briefing for Epson TM-m30. Fork of PrinterAutomation.

161 KiB

Languages

Python 97.8%

Shell 2.2%