Stats
Actions
Tags
From ait
Process files from Team Inbox (the user's input channel) into the PKA knowledge base. Extracts text, metadata, and tags, then indexes for search.
How this skill is triggered — by the user, by Claude, or both
Slash command
/ait:ait-ingestThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Process files and add them to the PKA SQLite knowledge base using the Python CLI.
Process files and add them to the PKA SQLite knowledge base using the Python CLI.
Before anything else, find where the AIT plugin is installed:
AIT_PLUGIN_DIR=""
for d in ~/.claude/plugins/marketplaces/*/; do
if [ -f "$d/.claude-plugin/plugin.json" ] && grep -q '"ait"' "$d/.claude-plugin/plugin.json" 2>/dev/null; then
AIT_PLUGIN_DIR="$d"
break
fi
done
[ -z "$AIT_PLUGIN_DIR" ] && [ -f ".claude-plugin/plugin.json" ] && AIT_PLUGIN_DIR="$(pwd)"
[ -z "$AIT_PLUGIN_DIR" ] && [ -d "pkb" ] && AIT_PLUGIN_DIR="$(pwd)"
echo "AIT_PLUGIN_DIR=$AIT_PLUGIN_DIR"
All python -m pkb.cli commands below must be run with PYTHONPATH="$AIT_PLUGIN_DIR" prepended. Also ensure click is available:
python3 -c "import click" 2>/dev/null || pip install click>=8.0 2>/dev/null || pip3 install click>=8.0 2>/dev/null
If $ARGUMENTS is provided, process that specific file or directory. Otherwise, process all files in Team Inbox/.
Look for kb/pka.db in the project root. If it doesn't exist, tell the user to run /ait-setup first.
Use the Python CLI for all database operations. Never use raw sqlite3 commands with string interpolation — file content and titles commonly contain characters that break SQL.
To add a single file:
PYTHONPATH="$AIT_PLUGIN_DIR" python3 -m pkb.cli add --db kb/pka.db "path/to/file"
To ingest an entire directory (default: Team Inbox/):
PYTHONPATH="$AIT_PLUGIN_DIR" python3 -m pkb.cli ingest --db kb/pka.db "Team Inbox/"
To ingest recursively:
PYTHONPATH="$AIT_PLUGIN_DIR" python3 -m pkb.cli ingest --db kb/pka.db -r "path/to/directory"
To add tags during ingestion:
PYTHONPATH="$AIT_PLUGIN_DIR" python3 -m pkb.cli add --db kb/pka.db -t "tag1" -t "tag2" "path/to/file"
The CLI automatically:
Display a summary:
Ingestion complete
- Files processed: {N}
- New items added: {N}
- Duplicates skipped: {N}
Search the knowledge base with:
python -m pkb.cli search --db kb/pka.db "your query"
npx claudepluginhub rappdw/aitIngests specific markdown files into Knowledge Base: distills content to KB format, creates files with frontmatter, registers in CLAUDE.md. Targeted alternative to /kb-absorb.
Ingests, indexes, imports, or processes files and folders into the Retriever workspace. Routes between ingest and ingest-production, bootstraps the workspace if needed, and summarizes results.
Bulk imports knowledge from files, directories, or URLs into structured backlogs, or captures a single document with a 5-section template (claims, worth-keeping, contested, action, reaction).