# Pkc Search

> Full-text search over the knowledge bundle (AND terms, type filters). Use when finding decisions, meetings, or features by keyword.

- Skill: `spillwavesolutions/pkc-search` (Agent Skill)
- Install (CLI): `npx skillmds@latest add spillwavesolutions/pkc-search`
- Raw SKILL.md: https://api.skillmd.com/api/skills/spillwavesolutions/pkc-search/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: spillwavesolutions (https://skillmd.com/u/spillwavesolutions)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/spillwavesolutions/pkc-search

---


# PKC Search

```bash
python3 "${CLAUDE_PLUGIN_ROOT}/scripts/pkc_search.py" "JWT refresh" --repo .
python3 "${CLAUDE_PLUGIN_ROOT}/scripts/pkc_search.py" JWT --type DecisionRecord,Feature --json
python3 "${CLAUDE_PLUGIN_ROOT}/scripts/pkc_search.py" JWT --engine index
python3 "${CLAUDE_PLUGIN_ROOT}/scripts/pkc_search.py" JWT --no-index --rg
python3 "${CLAUDE_PLUGIN_ROOT}/scripts/pkc_search.py" JWT --engine scan
```

AND semantics across terms. Scores title > description > tags > body.

Ladder: SQLite index → ripgrep → full scan. Ranking stays in Python, so
`--engine scan` and the index path return the same scores. `--engine fts`
uses FTS5 MATCH (prefix tokens; not score-identical). Missing index or rg
is not an error. See `/pkc-index` and `/pkc-setup`.

After hits, the parent should spawn `/pkc-retrieve` rather than packing
inline. Offer `/pkc-context --tiny` only when a deterministic pack is the
goal (verification, mermaid), not Q&A retrieval.