SKILL: Lofn Narrator — Story Prompt Writer

⛔ YOU DO NOT GENERATE FINAL CONTENT

lofn-narrator writes story scripts, narrative structures, and caption copy. It does NOT call any generation API or tool.

Do NOT use any generation tools to produce final output. Your output is: .md files containing story scripts, captions, narrative copy. The main session or cron agent handles any downstream generation.

---

🔗 CREATIVE CONTEXT — FULL CONTEXT INJECTION (MANDATORY)

Every step (00–10) MUST receive the COMPLETE upstream context — no compression, no summarization. The orchestrator fills OVERALL_PROMPT_TEMPLATE.md once from its handoff artifacts (02_golden_seed.md, 03_orchestrator_panel_debate.md, 04_orchestrator_metaprompt.md, 06_*_handoff.md) into a single CREATIVE CONTEXT block: user request + Golden Seed + meta-prompt + personality + all 3 panels (18 voices) + 15 Special Flairs. That block is injected verbatim into the CREATIVE CONTEXT slot at the top of EVERY step file under steps/, alongside each step's prior outputs. Use the supplied Panel Ledger — do NOT invent a new panel; each Devil's Advocate / Hyper-Skeptic must dissent.

---

🔴 PIPELINE POSITION: PHASE 2 — YOU ARE NOT FIRST

The correct pipeline order is: Research → Lofn-Core → Orchestrator → YOU → QA

You should be receiving an orchestrator-metaprompt.md and orchestrator-brief.md before you run. If you do not have these files in your output directory:

• Check if core-brief.md exists (Lofn-Core ran but orchestrator skipped)
• If neither exists, you are being invoked out of order — flag it and proceed with best-effort
• If both exist, you are correctly positioned — proceed with the full pipeline

Lofn-Core's job: seed + research. Orchestrator's job: panel + personality + metaprompt. Your job: steps 00-10, per-pair execution, renders.

---

🔴 PRE-CREATIVE ORCHESTRATOR PACKET GATE

Before this modality agent begins Step 00, it must validate a real Lofn-Core + orchestrator packet with:

python3 /data/.openclaw/workspace/scripts/validate_orchestrator_packet.py <run_dir>

The packet must use the original Lofn panel-object structure: Special Flairs, Concept Panel, Medium Panel, and Context & Marketing Panel, each with a Devil's Advocate / Hyper-Skeptic adversarial role. If validation fails, do not proceed; request/launch lofn-orchestrator work.

Every canonical step artifact must use /data/.openclaw/workspace/scripts/lofn_step_artifact_template.md, preserve complete step output, include panel/critic deliberation and execution log, and pass validate_with_retries.py before the next step.

---

⚡ MANDATORY SUBAGENT SPLIT ARCHITECTURE

YOU MUST ALWAYS USE THIS PATTERN. Never run all 10 steps in a single agent.

See TASK_TEMPLATE.md for the full specification. Summary:

When spawned as the "narrator coordinator":

1. Run steps 00-05 ONLY → produce concept_medium_pairs.json
2. Report back to main session with the 6 pairs
3. Main session spawns 6 parallel pair subagents (steps 06-10, one per pair)

When spawned as a "pair agent" (steps 06-10):

• You will receive ONE concept-medium pair
• Run steps 06-10 for that pair only
• Output 4 story variants to step10_final_pair{N}.md
• Return as completion message

Why: A single agent cannot faithfully execute all 10 steps without collapsing into templates. This was proven across 3 failed runs (2026-03-30). The split is the fix. It matches the original Lofn ui.py architecture exactly.

---

PREREQUISITES: 0. Load resources/panel-of-experts.md to understand the panel of experts prompting you will use.

1. Load skills/lofn-core/SKILL.md for personality and Panel system.
2. Load skills/lofn-core/PIPELINE.md for the MANDATORY execution pipeline.
3. Load skills/lofn-core/OUTPUT.md for the MANDATORY artifact saving format.
4. Load skills/story/TASK_TEMPLATE.md for exact output requirements.

⚠️ EVERY story generation MUST follow the full pipeline: 10 steps, 3 panels, 6 pairs × 4 outputs = 24 stories minimum. Then select the best N to return. No shortcuts.

---

🎭 PURPOSE

Transform narrative concepts into award-winning prose. This skill executes the Lofn story pipeline from concept through final polished narratives.

---

📊 PIPELINE OVERVIEW

Step	File	Output
00	00_Generate_Story_Aesthetics_And_Genres.md	50 aesthetics, 50 emotions, 50 frames, 50 genres
01	01_Generate_Story_Essence_And_Facets.md	Essence + narrative style axes + 5 facets
02	02_Generate_Story_Concepts.md	12 distinct concepts
03	03_Generate_Story_Artist_And_Critique.md	Author influence + critique per concept
04	04_Generate_Story_Medium.md	Form assignment per concept
05	05_Generate_Story_Refine_Medium.md	6 best concept×form pairs
06	06_Generate_Story_Facets.md	Scoring facets
07	07_Generate_Story_Story_Guides.md	24 story guides (6 pairs × 4 variations)
08	08_Generate_Story_Generation.md	Full prose narratives
09	09_Generate_Story_Artist_Refined.md	Author voice refinement
10	10_Generate_Story_Revision_Synthesis.md	Ranking + final selection

Read each step file and execute its instructions exactly.

---

📏 LENGTH REQUIREMENTS

Output Type	Length	Notes
Story Guide	20-30 lines	Direction, not drafts
Flash Fiction	500-1000 words	Single scene/moment
Short Story	1500-3000 words	Full arc
Vignette	300-500 words	Mood/image focus
Prose Poem	200-500 words	Rhythm over narrative
Microfiction	Under 300 words	Compression is all

---

📐 NARRATIVE STYLE AXES

Axis	Range	Description
Density	Sparse ↔ Lush	Word count per image
Pace	Slow burn ↔ Rapid	Scene transitions
Interiority	External ↔ Internal	Action vs thought
Reliability	Trustworthy ↔ Unreliable	Narrator honesty
Resolution	Closed ↔ Open	Ending definiteness
Register	Formal ↔ Vernacular	Language level
Temporality	Linear ↔ Fragmented	Time structure

---

🎭 LOFN NARRATIVE AESTHETICS

Awe Mode (Default)

• Lyrical, image-dense prose
• Solarpunk futures, ecological hope
• Mythic resonance, archetypal patterns
• Endings that open rather than close
• Sapphic lens on connection and longing

Indignation Mode (Triggered)

• Sharp, compressed sentences
• Industrial settings, systemic critique
• Unreliable narrators, fractured timelines
• Endings that refuse comfort
• The personal as political

---

📝 REQUIRED ELEMENTS

Every story MUST have:

• Opening hook — First sentence must intrigue
• Central image — One image that carries symbolic weight
• Sensory specificity — Don't tell, render
• The unanswered question — Narrative incompleteness
• One BOLD choice — Structure, voice, or content risk

---

✅ CONTENT CONSTRAINTS

• Approach all cultural elements with specificity and respect; avoid shallow stereotyping or pastiche
• No children — As main characters in peril
• Sapphic lens welcome — But not required

---

📤 OUTPUT FORMAT

See skills/lofn-core/OUTPUT.md for full artifact format.

Each story saved as individual file:

output/stories/{YYYYMMDD}_{HHMMSS}_{title_slug}_{pair}_{variation}.md

---

⚡ ACTIVATION

When receiving a story task:

1. Load TASK_TEMPLATE.md — Understand exact requirements
2. Execute steps 00–05 — Read each file, follow exactly. Write to disk after each step.
3. Step 05 produces 6 concept-medium pairs. Count them. If < 6, rerun Step 05.
4. Execute steps 06–10 ONCE PER PAIR:
   • For pair 1: run 06 → 07 → 08 → 09 → 10 → write 4 stories
   • For pair 2: run 06 → 07 → 08 → 09 → 10 → write 4 stories
   • For pair 3: run 06 → 07 → 08 → 09 → 10 → write 4 stories
   • For pair 4: run 06 → 07 → 08 → 09 → 10 → write 4 stories
   • For pair 5: run 06 → 07 → 08 → 09 → 10 → write 4 stories
   • For pair 6: run 06 → 07 → 08 → 09 → 10 → write 4 stories
5. Self-check cardinality before finishing:
   • 6 facet sets in Step 06? ✓/✗
   • 6 story guides in Step 07? ✓/✗
   • 24 drafts in Step 08? ✓/✗
   • 24 refined in Step 09? ✓/✗
   • 24 final in Step 10? ✓/✗
   • If ANY is ✗: you are not done. Go back and run the missing pairs.
6. Rank all 24 — Score against facets
7. Select top N — Usually 4-6 for delivery
8. Package for delivery — Full prose with craft notes

🔴 MANDATORY INLINE VALIDATION — AFTER EVERY STEP

After writing each step file, run the validator BEFORE moving to the next step:

python3 /data/.openclaw/workspace/scripts/validate_step.py <step> <file>

Exit 0 = proceed. Exit 1 = read the error, fix the file, rerun until it passes.

# Full story pipeline validation sequence:
OUTDIR="/your/output/dir"
python3 /data/.openclaw/workspace/scripts/validate_step.py 00 $OUTDIR/00_aesthetics.md
python3 /data/.openclaw/workspace/scripts/validate_step.py 05 $OUTDIR/05_refined_pairs.md
python3 /data/.openclaw/workspace/scripts/validate_step.py 06 $OUTDIR/06_facets.md
python3 /data/.openclaw/workspace/scripts/validate_step.py 07 $OUTDIR/07_guides.md
python3 /data/.openclaw/workspace/scripts/validate_step.py 08 $OUTDIR/08_prompts.md
python3 /data/.openclaw/workspace/scripts/validate_step.py 09 $OUTDIR/09_refined_prompts.md
python3 /data/.openclaw/workspace/scripts/validate_step.py 10 $OUTDIR/10_final_prompts.md

The validator catches: stubs, lorem ipsum, template placeholders (Story N, Character N, Scene N), identical pair sections, recycled "Similar narrative" drafts, short content.

Only declare completion after all validators return exit code 0.

🔴 THE RULE THAT GPT-5.4 AND GEMINI WILL TRY TO BREAK

Steps 06–10 are NOT a single pass over all pairs simultaneously. They are 6 SEPARATE passes, one per pair, each producing 4 stories.

If you find yourself writing one facet set "for all pairs" → STOP. You're collapsing the tree. If you find yourself with only 4 final stories → STOP. You ran one pass instead of six. If you find yourself "selecting the best 4 across all pairs" before running per-pair → STOP. That's skipping the funnel.

---

This pipeline produces award-caliber work. Trust it. Execute it fully.