Skip to content

v0.1.5 — Briefing severity fix, properly

Choose a tag to compare

@ThinkCode ThinkCode released this 20 Oct 04:50
· 5 commits to main since this release

v0.1.5 — Briefing severity fix, properly

Follow-up to v0.1.4's briefing rendering fix, with two corrections found in a live run:

  • Prompt templates no longer get copied literally. Apple Foundation Models reproduced the prompt's placeholder shape (• [HIGH] finding — …) instead of filling in real findings. The prompts now show a concrete example instead of a placeholder pattern, so every finding names the actual item and its size. A storage-prompt version is baked into the per-scan section cache, so prompt improvements take effect on the next briefing instead of being blocked by stale cached text.
  • "Needs attention" now shows exactly the tagged findings — untagged prose (section summaries, clean-network bullets) is filtered out, and every remaining item carries its red/amber severity tag, bullet or not.

Plus everything from v0.1.4: real version number in the About panel, severity tags colored on every line.

SHA-256

88e2074199b98b03634382ad0a3a90ba7c164b0a202b2f58f60f7fae573df960  System_Cockpit_0.1.5_aarch64.dmg
b9b1dd34e38ad31fc775a57f886cac6b616b21bc4c97a570db48e3ddf058d040  System_Cockpit_0.1.5_x64.dmg

Both DMGs are notarized and stapled (Developer ID: BYTEXIO LLC, 5TN66T2V5U).