Overcast Init

overcast-init

One-time setup for overcast.

1. Install the CLI — pi install npm:@kdrrr/overcast (inside pi) or

npm i -g @kdrrr/overcast for the standalone binary.

1. Install bundled skills for this agent — for Codex, Cursor, and other

non-Claude harnesses, copy the shipped skills into the agent's skills directory explicitly: ``bash overcast skills install --dest ~/.codex/skills ` Bare overcast skills install remains the Claude Code default (~/.claude/skills), and --harness claude-code` is the explicit Claude target.

1. Install/update tinycloud — the default perception backend. Get the latest

(npm i -g @cloudglue/tinycloud@0.3.8 then tinycloud install --latest, or tinycloud update). The face + index verbs need tinycloud ≥ 0.3.4, and overcast currently recommends 0.3.8 (the image see/extract verbs behind the opt-in see:tinycloud provider need ≥ 0.3.7); override the invocation with OVERCAST_TINYCLOUD_CMD if it isn't on PATH.

1. Verify — overcast doctor --json (pi pinned, ffmpeg/ffprobe runnable,

Cloudglue key, tinycloud CLI + version, optional uv/visual-db readiness).

1. Cloudglue key — the default watch/listen/face/index providers

reach Cloudglue via the tinycloud CLI; configure it (tinycloud setup cloudglue) or export CLOUDGLUE_API_KEY.

1. Provider profile setup — choose reusable providers once per profile, not

once per case. Always preview before applying: ``bash overcast provider setup show --profile default --json overcast provider setup plan --preset cloudglue --profile default --json overcast provider setup apply --preset cloudglue --profile default --yes --json overcast doctor --profile default --json `` Optional presets/choices:

• cloudglue for tinycloud watch/listen/face plus built-in ffmpeg enhance.
• fal for see/enhance with FAL_KEY.
• hf for see/enhance with HF_TOKEN.
• elevenlabs for listen/enhance with ELEVENLABS_API_KEY.
• owl-local for OWLv2 open-vocabulary object detection.
• see:tinycloud (choice, --verb see --choice tinycloud) for Cloudglue

file-level image analysis via tinycloud see/extract (needs tinycloud ≥ 0.3.7; --detect facts are boxless — no crop).

• deepface-local for local face detect/match through DeepFace.

1. Optional visual DB setup — prepare visual DB Python once per

checkout/machine. DeepFace can be selected as a profile provider for the face verb, while image/face DBs are still case-owned local indexes: ``bash scripts/visual-db-uv.sh --face overcast doctor --json overcast provider setup apply --verb face --choice deepface-local --profile default --yes --json overcast index create logos --type image-ransac --local --json overcast index create localfaces --type deepface-local --local --json ``

1. Case setup later — use the main overcast skill per investigation to run

case setup, select targets/sources/indexes, and optionally set case-level automation such as --auto-sense, --auto-index-new, and --findings (defaults to suggest — score/text triggers auto-suggest leads; review is the legacy text-only mode, off disables).

Then use the overcast skill to drive the verbs.

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: kdr
• Source: kdr/overcast
• License: Apache-2.0
• Homepage: https://overcast.video

Install and usage instructions live in the source repository linked above.