| name | tools-skill |
| description | Maintain deterministic performance, screenshot, fleet, geometry, asset, and release verification tools. |
claude-of-tanks / tools
Purpose
Provide reproducible evidence for game performance, rendering, tank fidelity,
asset provenance, and public builds.
attribution-audit.mjs enforces the repository-wide Kevin B. Liu notice,
named authorship for every playable model, and exact records for tracked
external model files.
Mental model & key files
Performance probes drive the browser and record JSON; fleet/geometry tools audit
authored tanks; screenshot/visual tools stage canonical views; strip/release
tools enforce public asset boundaries.
Patterns to follow / invariants
Pin URL, flags, roster, timings, and output path. Make gates fail visibly and
avoid editing generated evidence manually. Combat-anatomy generation always
uses ALL_TANK_IDS; donor/retired spec rows are not part of the playable gate.
Common tasks → first action
Read the tool's CLI/help and its current evidence doc, run a baseline, then
compare the same scenario after changes. Multiplayer release checks include the
two-player persistent-room soak, human 2v2 (npm run test:net:four), and full
human 7v7 capacity (npm run test:net:seven) browser paths. Visual combat
certification is the separate npm run test:net:seven:live gate: two real 7v7
matches render the host and an impaired remote client while all fourteen tanks
move, fire, deal damage, and report transport/prediction/frame/shadow health.
Tank work must run npm run tank:anatomy:update before asset/release checks;
the update refreshes the receipt map and only the three fleet technical views,
preserving unrelated garage/top/side/markings assets.
Gotchas
Many tmp-* tools and .qa-dev/ outputs are transient and must not be staged.
Own and stop every dev server/browser process you start.
