lit-capturelisted

<!-- Copyright (c) 2026 JG Systems Consulting Ltd. Source: https://github.com/jgsystemsconsulting/jgs-lit-memory. See LICENSE. --> <!-- SPDX-License-Identifier: MIT --> # lit-capture Run this as a procedure. The corpus lives in `.lit/` at the root of the current project repo and is git-tracked there. The script `lit_fetch.py` is the only component that touches the network, and it talks to OpenAlex only. Everything else is agent reading and writing. Invoke the script by its installed path so no PATH setup is needed: ```bash python "$HOME/.zcode/skills/lit-capture/lit_fetch.py" --status ``` Common flags: `--dir <path>` (corpus root, default `.lit`; always run from the repo root), `--api-key <key>` (default env `OPENALEX_API_KEY`). ## Prerequisites - Python 3.9+ (standard library only; no pip packages) reachable as `python`. - Network access to `api.openalex.org`. Singleton DOI and W-id lookups are free; title searches, batches, and inbox triage draw on a daily budget, so set `OPENALEX_API_KEY` (from OpenAlex) for the full budget. - The script ships inside this skill folder. The default install is ZCode (`~/.zcode/skills/lit-capture/lit_fetch.py`); on other hosts substitute that host's installed path (for example `~/.claude/skills/jgs/lit-capture/lit_fetch.py`). ## When to use A paper enters the conversation with enough identity to fetch (DOI, arXiv id, W-id, or exact title plus author and/or year) and looks worth keeping. Capture takes under a minute and neve