1.8 KiB
1.8 KiB
CLAUDE.md
Optimizer: the optimisation of the AIMsight detectors and of the reading tools, from one analysis of labeling/auto, and nothing else (README.md). The workflow: .claude/skills/optimize/SKILL.md (/optimize), launched from the root of Optimizer with the agents plugin (--plugin-dir agents); its agents agents:upgrader, agents:validator (the detectors) and agents:tool-builder (the tools); the benchmark detectors/bench/.
- The links, out of git:
detectors/→~/detectorsandtools/→~/tools(the session commits there);data/→~/aimsight-data/dataandlabeling/→~/labeling(read-only);agents/→~/agents(the plugin, loaded read-only). - The sessions:
sessions/<analysis>_<YYYYMMDDTHHMMZ>/; Diagnostic-App reads their summary (summary.md). - The workflow stays generic: the skill describes only the workflow (analysis → detectors: upgrader → validator; tools: tool-builder → summary); what is specific to a caller (its paths, its git rules,
Caller's context) arrives in the request. - Check a change of the skill:
env -u ANTHROPIC_API_KEY claude -p "/optimize" --plugin-dir agents --model haikumust say the inputs are missing and stop without touchingdetectors/ortools/; then a real session through the app's button. archive/: the former batch and its scanner, stopped; do not run them again.
Rules:
- The agents run on the Claude subscription, never on an API key (
ANTHROPIC_API_KEYremoved from the launch). - Every time is UTC.
- Read-only on
data/andlabeling/; never copy the metric identifiers (SIM_CCID,ip*); never read.env. - The project uses the labels of
labeling/autoonly;labeling/manualis the user's alone. - No rule specific to a station, a target or a date.
- Commit by path, never
git add -A. - Language: English.