OEJTS Personality Tuner

Use this skill to run the OEJTS 1.2 assessment, compute type results, and convert the outcome into concrete assistant behavior preferences.

Quick Start

• Show questionnaire:
• python3 scripts/oejts_tuner.py template --format markdown
• Score answers:
• python3 scripts/oejts_tuner.py score --answers-json '{"Q1":3,...,"Q32":4}'
• Generate proposed updates for workspace files:
• python3 scripts/oejts_tuner.py apply --workspace /path/to/workspace --answers-json '{...}' --dry-run
• Apply updates:
• python3 scripts/oejts_tuner.py apply --workspace /path/to/workspace --answers-json '{...}'

Workflow

1. Administer OEJTS 1.2 using template output from the script.
2. Collect 1–5 answers for Q1..Q32.
3. Run score to compute IE, SN, FT, JP, type letters, and confidence.
4. Run apply --dry-run to preview USER.md/SOUL.md changes.
5. Confirm with user, then run apply without --dry-run.

Rules

• Use OEJTS type as a preference signal, not fixed identity.
• Prefer adjustable behavior vectors over stereotypes.
• Keep user override available (user feedback beats test output).
• Preserve existing USER.md/SOUL.md content; append/update only the managed blocks.

Managed Blocks

This skill writes only between these markers:

• USER.md:
• SOUL.md:

If blocks do not exist, they are appended.

References

• references/oejts-1.2.md — canonical OEJTS items and scoring equations.
• references/behavior-mapping.md — type-to-behavior mapping and adaptive guidance.