a69b9d9160
Permission-first AI agent runtime for Google Workspace. Ports the LARC/OpenClaw governance model (disclosure chain, execution gates, queue/ingress) to Gmail, Calendar, Drive, Sheets, Tasks, and People APIs with Claude Code as the execution engine. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
123 lines
3 KiB
Bash
123 lines
3 KiB
Bash
#!/usr/bin/env bash
|
|
# GARC kg.sh — Knowledge graph via Google Docs
|
|
# Google Docs replaces Lark Wiki as the knowledge graph surface
|
|
|
|
GARC_KG_CACHE="${GARC_CACHE_DIR:-${HOME}/.garc/cache}/knowledge-graph.json"
|
|
|
|
garc_kg() {
|
|
local subcommand="${1:-help}"
|
|
shift || true
|
|
|
|
case "${subcommand}" in
|
|
build) garc_kg_build "$@" ;;
|
|
query) garc_kg_query "$@" ;;
|
|
show) garc_kg_show "$@" ;;
|
|
*)
|
|
echo "Usage: garc kg <build|query|show>"
|
|
return 1
|
|
;;
|
|
esac
|
|
}
|
|
|
|
# garc kg build
|
|
# Crawls Google Drive folder and builds knowledge graph from Docs
|
|
garc_kg_build() {
|
|
local folder_id="${GARC_DRIVE_FOLDER_ID:-}"
|
|
|
|
if [[ -z "${folder_id}" ]]; then
|
|
echo "Error: GARC_DRIVE_FOLDER_ID not set" >&2
|
|
return 1
|
|
fi
|
|
|
|
echo "Building knowledge graph from Google Drive folder: ${folder_id}"
|
|
|
|
python3 "${GARC_DIR}/scripts/garc-drive-helper.py" kg-build \
|
|
--folder-id "${folder_id}" \
|
|
--output "${GARC_KG_CACHE}"
|
|
|
|
echo "✅ Knowledge graph built: ${GARC_KG_CACHE}"
|
|
}
|
|
|
|
# garc kg query "<concept>"
|
|
garc_kg_query() {
|
|
local query="$*"
|
|
|
|
if [[ -z "${query}" ]]; then
|
|
echo "Usage: garc kg query \"<concept>\""
|
|
return 1
|
|
fi
|
|
|
|
if [[ ! -f "${GARC_KG_CACHE}" ]]; then
|
|
echo "Knowledge graph not built. Run: garc kg build"
|
|
return 1
|
|
fi
|
|
|
|
python3 -c "
|
|
import json, sys
|
|
query = '${query}'.lower()
|
|
with open('${GARC_KG_CACHE}') as f:
|
|
kg = json.load(f)
|
|
|
|
matches = []
|
|
for node in kg.get('nodes', []):
|
|
name = node.get('title', '').lower()
|
|
content = node.get('content_preview', '').lower()
|
|
if query in name or query in content:
|
|
matches.append(node)
|
|
|
|
if not matches:
|
|
print(f'No results for: ${query}')
|
|
sys.exit(0)
|
|
|
|
print(f'Results for \"{query}\" ({len(matches)} matches):')
|
|
for m in matches[:10]:
|
|
print(f' - [{m.get(\"doc_id\",\"\")}] {m.get(\"title\",\"\")}')
|
|
if m.get('content_preview'):
|
|
preview = m['content_preview'][:100].replace('\n', ' ')
|
|
print(f' {preview}...')
|
|
links = m.get('links', [])
|
|
if links:
|
|
print(f' Links: {len(links)} documents')
|
|
"
|
|
}
|
|
|
|
# garc kg show <doc_id>
|
|
garc_kg_show() {
|
|
local doc_id="${1:-}"
|
|
|
|
if [[ -z "${doc_id}" ]]; then
|
|
echo "Usage: garc kg show <doc_id>"
|
|
return 1
|
|
fi
|
|
|
|
if [[ ! -f "${GARC_KG_CACHE}" ]]; then
|
|
echo "Knowledge graph not built. Run: garc kg build"
|
|
return 1
|
|
fi
|
|
|
|
python3 -c "
|
|
import json
|
|
doc_id = '${doc_id}'
|
|
with open('${GARC_KG_CACHE}') as f:
|
|
kg = json.load(f)
|
|
|
|
for node in kg.get('nodes', []):
|
|
if node.get('doc_id') == doc_id:
|
|
print(f'Title: {node.get(\"title\", \"\")}')
|
|
print(f'Doc ID: {doc_id}')
|
|
print(f'Type: {node.get(\"mime_type\", \"\")}')
|
|
print(f'Modified: {node.get(\"modified_time\", \"\")}')
|
|
print()
|
|
print('Content preview:')
|
|
print(node.get('content_preview', '(none)'))
|
|
print()
|
|
links = node.get('links', [])
|
|
if links:
|
|
print(f'Links ({len(links)}):')
|
|
for link in links:
|
|
print(f' -> {link}')
|
|
break
|
|
else:
|
|
print(f'Document {doc_id} not found in knowledge graph')
|
|
"
|
|
}
|