ckb/.agents/modules/capture/module.yaml
Michał Kopeć dad2264a65 Template 1.26.0: release 3 batch 3, capture module, skill evals, multi-agent docs
- F11 new opt-in module `capture` (skill ckb-capture, capture_url.py):
  stdlib fetch with readable-text extraction, Wayback only over a failed
  or thinner live page, --from-file for the agent's own fetch, PDFs saved
  as files, --media via yt-dlp with consent, untrusted: true marking,
  dedup at capture time; dedup_inbox.py now compares source_url (D-0017)
- F22 tests/run_evals.py (list/prepare/check/validate) and five
  structural eval cases; reference run for ckb-decide/record
- F18 manual section with untested setups for Codex CLI, Cursor,
  OpenCode and Gemini CLI, plus the Claude Code-only feature table;
  AGENTS.md tells agents without skill loading to read SKILL.md

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-02 20:44:05 +01:00

30 lines
949 B
YAML

# Cascade KB module manifest — read by `ckb-module` on install/uninstall.
# See docs/decisions/0017-capture-module.md.
name: capture
title: URL and media capture
version: 1.0.0
requires_kb_schema: "1.5"
summary: >
Turns a link or a video into a dated Markdown snapshot in raw/inbox/, with
the source URL and fetch date in frontmatter, so web material goes through
the same ingest, check-in and fingerprints as any other source.
skills:
- ckb-capture
agents_fragment: fragments/agents-routing.md
gitignore_fragment: fragments/gitignore.snippet
schema_additions:
types:
- capture # a raw/inbox/ snapshot; becomes an archived source, never a wiki page
fields: []
depends_on:
- name: yt-dlp
scope: media-only
check: "command -v yt-dlp"
optional: true
note: >
Needed only for video and podcast transcripts. Pages and PDFs are fetched
with the Python standard library. Never installed without asking.