git:20260814.24aaeeb to git:20260814.28cab41

9 added, 9 removed. Audit A to A.

---

name: script-adaptation

description: Adapt a historical-drama Work into a screen Script. Use when designing the audiovisual dramatic structure, main and secondary lines, character arcs, pacing, escalation, climax, or short-form series shape.

---

# Script Adaptation

- Preserve the Work's dramatic truth while translating it into observable screen action. Define the main line, necessary secondary lines, character arcs, pacing, escalation, climax, and short-form structure without replacing the Work's historical or creative boundaries.

+ Convert an approved Work into a screen-adaptable formal Script state that can govern later Episode development. Inherit the Work's story foundation while designing causal dramatic progression, visible character arcs, information reveal, episode architecture, short-form pacing, and playable storytelling; do not redesign the Work or prematurely write Scene/Shot detail.

## Creative Lifecycle

### 1. Understand Goal

- Clarify whether the request creates a new Script or revises an existing one, the target Work, requested adaptation scope and outcome, and explicit format, length, audience, tone, or continuity constraints.

+ Clarify whether the request creates a new Script or revises an existing one, the target Work, requested adaptation scope and outcome, and explicit format, episode count or duration, audience, tone, or continuity constraints. If the request actually changes the approved protagonist, premise, historical boundary, major ending, or other Work foundation, identify the upstream issue instead of silently repairing it in Script.

### 2. Gather Context

Assess context sufficiency before planning. Continue when the relevant Work and adaptation constraints are supplied. Use `work.get_work` for a known `workId`; use `script.get_script` for a known `scriptId`; otherwise use `script.list_scripts` under the known Work and judge the structured candidates. Use `context.build_context` only when required Work or existing Script context was not supplied.

- Use adequate research context first. When a consequential adaptation decision depends on an unresolved historical claim, use `research.verify_claim`. If broader evidence is required, formulate a focused research question and stop before planning or persistence so the Agent or Host can choose an existing research capability. Do not repeat research for settled context. If the governing Work, essential constraints, or consequential evidence cannot be obtained or conflict, state the blocker and do not draft or persist.

+ Use the Work's approved evidence boundary and supplied research context first. Use `research.verify_claim` only for a consequential historical claim newly introduced or materially reinterpreted by the adaptation; do not repeat research for facts already approved in Work. If broader evidence is required, formulate a focused research question and stop before planning or persistence so the Agent or Host can choose an existing research capability. If the governing Work, essential constraints, or consequential evidence cannot be obtained or conflict, state the blocker and do not draft or persist.

### 3. Plan

- Create an internal Script plan stating the adaptation goal, Work facts and character truths that must be inherited, constraints that cannot be violated, the intended audiovisual structure and progression, the main/secondary lines, arcs, pacing, escalation, climax, and short-form shape required in the draft, and unresolved questions that must be settled first. Keep it in Agent Run Context or temporary working state. Do not call `script.create_script` or `script.save_script` to store it.

+ Before drafting, read [Professional Script Planning](references/planning.md) and apply it. Create an internal adaptation contract that records the Work premise, protagonist, major arcs, central conflict, stakes, thematic question, historical boundary, major turns, ending, format, audience, and tone that must be inherited. Plan a causal main line, only necessary secondary lines, screen-visible arc progression, escalating conflict, information reveal, episode architecture, short-form pacing, climax/payoff, screenability, and dialogue strategy. Resolve open structural questions before drafting. Keep the plan in Agent Run Context or temporary working state. Do not call `script.create_script` or `script.save_script` to store it.

### 4. Execute Draft

- Execute the plan as a complete candidate formal Script state. Prefer observable action over explanatory prose and preserve continuity with the Work. The draft must be detailed enough for Script review and later Episode development; it must not be a plot-summary fragment, placeholder, test content, or scratchpad. Do not persist a partial draft.

+ Execute the plan as a complete candidate formal Script state. Translate abstract arcs and history into observable action, behavior, dialogue, visual information, discovery, choice, reaction, and consequence. Make each structural segment and proposed Episode unit change story state and advance the main line; use dialogue with dramatic purpose rather than textbook exposition. The result must let Episode development understand the whole-series drive, progression, arcs, episode jobs, turns, climax, ending, and historical constraints without reinventing them. It must not be a plot summary, event list, mechanical episode split, placeholder, test content, scratchpad, detailed Scene script, or camera plan. Do not persist a partial draft.

### 5. Review

- Review the complete draft before any write. Critical checks are: main and necessary secondary lines support the Work; character arcs, pacing, escalation, climax, and short-form structure are coherent; the result is screenable rather than merely explanatory; Work truth and consequential historical constraints remain intact; and no material continuity conflict is unresolved. Mark Review PASS only when every critical check passes; otherwise mark Review FAIL.

+ Before any write, read [Script Review and Revision](references/review.md) and apply the entire domain rubric to the complete draft. Critical checks cover Work fidelity, main and secondary lines, motivation and character arcs, causality and escalation, information reveal, episode architecture, pacing, screenability, dialogue, historical integrity, climax/payoff, ending fidelity, and downstream readiness. A plot summary, event list, mechanical episode split, unfilmable interior prose, or material Work drift is a critical failure. Mark Review PASS only when every critical check passes; otherwise mark Review FAIL.

### 6. Revise or Re-plan

- On Review FAIL, do not persist. Locally revise wording, isolated pacing, completeness, or minor continuity defects. Re-plan the current Script when the main line, character arc, overall structure, escalation, climax, or fidelity to the Work fails. After any revision or re-plan, review the complete draft again. A fix never goes directly to persistence without Review Again and PASS.

+ On Review FAIL, do not persist. Follow the defect routing in [Script Review and Revision](references/review.md): Locally revise dialogue, isolated beats, limited exposition, a small ordering problem, or minor continuity; Re-plan structurally when the main line, motivation, arc, conflict escalation, episode architecture, climax/payoff, or Work fidelity fails. If the defect originates in Work, mark it as an upstream Work issue rather than silently changing the foundation. After any revision or re-plan, review the complete draft again. A fix never goes directly to persistence without Review Again and PASS.

### 7. Persist

- No Review PASS means no create or save. Persist only when required context is sufficient, the plan and complete draft exist, all critical checks pass, and no historical or continuity conflict remains. Plan, draft reasoning, review notes, and revision notes remain Agent Run Context or temporary working state; do not put them in Script `content`. Persist only the reviewed formal result.

+ No Review PASS means no create or save. Persist only when required context is sufficient, the plan/adaptation contract and complete draft exist as a screen-adaptable formal Script state, all critical checks in the Script rubric pass, and no Work, historical, or continuity conflict remains. A plot summary, event list, few-line episode outline, passive main line, missing arc or escalation, absent climax, unworkable episode architecture, unfilmable prose, or material Work drift cannot pass. Plan, draft reasoning, rejected alternatives, review notes, and revision notes remain Agent Run Context or temporary working state; do not put them in Script `content`. Persist only the reviewed formal result.

Use `script.create_script` only for a genuinely new adaptation after producing the complete initial formal state needed by this Skill. A successful create is the normal first write and returns the stable ID; do not call `script.save_script` immediately afterward unless a concrete revision has actually occurred. Use `script.save_script` only to revise an already persisted Script because of a specific request, discovered error, upstream change, or necessary addition.

- Organize persistence as **Stable Envelope + Domain Content**. Keep the parent Work ID and Script title in the create envelope; use the stable Script ID and title for a revision. Put the reviewed audiovisual structure, lines, arcs, pacing, escalation, climax, and other formal adaptation facts in `content`. Do not move the parent Work ID or hide, duplicate, or rename envelope fields inside `content`. Treat the Tool catalog as the sole machine-schema source. Submit save as a full replacement formal state, never as a patch, scratchpad, stringified JSON, or routine follow-up to create. Use `context.refresh_context` only after a write makes current context stale. Do not create Episodes or call another Skill automatically.

+ Organize persistence as **Stable Envelope + Domain Content**. Keep the parent Work ID and Script title in the create envelope; use the stable Script ID and title for a revision. Put the reviewed adaptation contract, main/secondary lines, visible arcs, escalation, information reveal, episode architecture, pacing, climax, ending, screenability, and other formal adaptation facts in the open `content` object. These are creative content, not a request for new persistence fields. Do not move the parent Work ID or hide, duplicate, or rename envelope fields inside `content`. Treat the Tool catalog as the sole machine-schema source. Submit save as a full replacement formal state, never as a patch, scratchpad, stringified JSON, or routine follow-up to create. Use `context.refresh_context` only after a write makes current context stale. Script may plan Episode architecture, but do not create Episode entities, write detailed Scene dialogue/action, design cameras, or call another Skill automatically.