Ccstory

ccstory
Your AI coding-agent week, in plain English.
Reads local coding-agent session logs and writes a categorized recap with
active hours, costs, and a per-bucket narrative. This release bundles Claude
Code, OpenAI Codex, Google Antigravity, and the Grok CLI; the provider
registry keeps the same recap contract across them.

Sibling to ccusage:
ccusage tells you how much you spent · ccstory tells you what on.

Who this is for

- People who want to write a weekly status without scrolling scrollback.
- People who saw a ccusage number and want to know what kind of work those
  tokens went to.
- People who do a Sunday-night reflection on what they actually shipped.

Quick start

That's it. init is a one-time auto-categorize step that scans your
recent sessions; ccstory week produces the recap. Full report saves to
~/.ccstory/reports/recap-*.md.

The default Repo activity section always reads local git. When GitHub is
connected, it lightly enriches up to 10 active repos with GitHub metadata.
For a first run with no network access at all, use:

--no-artifacts alone disables ccstory's GitHub/PyPI lookups while keeping
the normal narrative flow, which may invoke one of your configured local
coding-agent CLIs.

Demo

The markdown report adds a header + 2-4 bullet points per bucket plus
per-session one-liners. Run with --llm-narrative to upgrade per-session
lines from the instant first/last-message fallback to LLM-polished prose:
Re-running upgrades retroactively. If you viewed a window in the
default (instant) mode first, re-running it with --llm-narrative upgrades
those cached fallbacks to polished summaries — so ccstory month
--llm-narrative polishes weeks you already skimmed. Already-polished
sessions are reused (no re-burn) while their exact bounded transcript
evidence,…