Validate the operation mode field in Stage 4 SG_Sum JSON against HARA scenarios.
npx skills add https://github.com/WSJ40/hara-mutiagent --skill hara-stage4r
Please help me install this Agent Skill. Skill: hara-stage4r Source: https://github.com/WSJ40/hara-mutiagent/tree/main/skills/hara-stage4r Command: npx skills add https://github.com/WSJ40/hara-mutiagent --skill hara-stage4r
It helps you verify that the model-filled “operation mode” in Stage 4 output is specific, non-placeholder, and consistent with the originating HARA scenario, without re-evaluating safety target semantics.
Ask the system to run Stage 4 machine validation and review the operation mode for output/<RUN_ID>_stage4_sg_sum.json, updating only the “操作模式” field if required.
High-intent search queries and answers about installing and using this skill.
To validate the operation mode in a Stage 4 HARA JSON, run Stage 4 machine checks on the output file to ensure the mode is present, specific, and not a placeholder. This verifies consistency with the originating HARA scenario.
A Stage 4R operation mode consistency review checks that the model-filled operation mode in a Stage 4 SG_Sum JSON is a specific mode name rather than a long scene description. It ensures alignment with the originating HARA scenario without re-evaluating safety target semantics.
No, when fixing the operation mode in Stage 4 JSON, you must only update the operation mode field. Safety targets, ASIL, safety state, and FTTI must remain consistent and untouched during the controlled repair process.
To repair a placeholder operation mode in a stage4_sg_sum.json file, run Stage 4 machine validation first, then perform a controlled fix that updates only the operation mode field. Consult Stage3 HARA references for the relevant MF_ID to ensure evidence-aligned consistency.
Automated Stage 4 operation mode repair is limited to updating only the operation mode field and cannot modify safety targets, ASIL, safety state, or FTTI. It requires prior machine validation and does not re-evaluate safety target semantics.
WSJ40Derive Stage 4 SG_Sum summaries from validated HARA results with tool-based grouping.
WSJ40Validate Stage 3B SEC ratings, FTTI consistency, and safety goal coverage.
zircoteAudit and repair filesystem-based AI memory systems.
WSJ40Derive Stage 1 function-level malfunction candidates from Stage 0 mappings.
