extract-flowchartslisted
Install: claude install-skill tamalCodes/birdsEye
# Extract flowcharts
> **Dormant as of the zero-token rewrite.** birdsEye's `/birdseye:map` no
> longer invokes this skill - the map is now a pure local AST parse (graphify)
> with no model calls. This file is kept for a future opt-in `--with-llm` mode.
This is what the Docs view is *for*. A reader who opens a feature's docs page
should not have to piece a flow together from prose - they should see it. This
skill reads what `extract-docs` already found and turns it into a small
step-by-step diagram: the same job you would do by hand if a user asked you to
read a module's docs and explain how it actually works.
Write `birdseye/.cache/flowcharts.json` and nothing else. Never modify the
user's source tree, and never write documentation into it.
## Non-negotiables
- **Ground every step in what the docs say.** If a doc does not describe it,
it does not go in the flow. This is the same rule `extract-docs` follows for
edges: a missing flowchart is fine, a wrong one is not.
- **A module with nothing to say gets no flowchart.** Reference tables,
guardrail-only files, and one-line READMEs do not describe a flow. Skip
them - do not invent steps to fill the gap.
- **Plain language.** Write every `label` and `detail` for someone who has
never opened this codebase and does not know what a "slice" or a "screen
navigator" is. Say what the user does and what happens, not what the code
does.
- **Small.** 3 to 12 steps. A flow longer than that is usually two flows -
pick t