| name | init |
| description | Initialize a Korean writing AGENTS.md by analyzing local blog, portfolio, report, message, or project samples with Python plus LLM interpretation. |
Write As Me KO Init
Use this skill when the user wants Codex or Claude Code to analyze their Korean
writing samples and build the writing setup for future drafting. The user should
not need to manually run each script.
Product Shape
This is an agent-led local workflow:
- The maintainer provides a repo, folder, or local writing samples.
- The agent runs deterministic Python analysis for counts, routes, sentence
length, repeated phrases, privacy notes, and baseline route signals.
- The agent reads representative samples and interprets writing habits with
the LLM.
- The agent updates
voice-profile.md conservatively.
- The agent exports a writing
AGENTS.md for reuse.
- The agent verifies the setup.
This is not model fine-tuning and not an AI-detector bypass workflow.
Steps
- Read
README.md, docs/product-goal.md, and docs/architecture.md.
- Locate samples:
- Prefer an explicit path from the user.
- Otherwise use
samples/.
- Treat
samples/private/ and *.local.md as sensitive local inputs.
- Run:
python -m scripts.init_writing_workspace --samples samples --repo-root .
- Read
_workspace/writing-init/llm-review.md.
- Inspect 3-10 representative samples when available. Do not copy private raw
sample text into git-tracked docs or exported AGENTS files.
- Edit
codex/skills/write-as-me-ko/references/voice-profile.md:
- Keep the Python quantitative signals.
- Add LLM-observed writing habits only when supported by samples.
- Mark thin or missing routes explicitly.
- Preserve uncertainty instead of pretending to know the user's voice.
- Re-export:
python -m scripts.export_agent_context --output dist/writing/AGENTS.md
- Verify:
npm run docs:check
Output
Report:
- sample source used
- files generated or updated
- what Python measured
- what the LLM interpreted
- confidence and missing sample routes
- verification result
Guardrails
- Do not commit private user writing.
- Do not claim perfect voice cloning.
- Do not claim AI detector bypass.
- Do not invent personal experiences, achievements, relationships, or metrics.
- If samples are too thin, produce a conservative writing profile and say so.