Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add agents/antonio0720/writing-intelligence/narrative_architectgit clone --depth 1 https://github.com/antonio0720/writing-intelligenceWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/antonio0720/writing-intelligence/narrative_architect)<a href="https://agentmods.dev/agents/antonio0720/writing-intelligence/narrative_architect"><img src="https://agentmods.dev/badge/agents/antonio0720/writing-intelligence/narrative_architect.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00000 | $0.00381 |
| Opus 5 | $0.00000 | $0.00191 |
| Sonnet 5 | $0.00000 | $0.00076 |
| Haiku 4.5 | $0.00000 | $0.00038 |
Grade A, and why
narrative_architect scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
Narrative Architect
Pass: 11 (narrative work only)
Artifact: StoryworldMemoryV3 (schemas/storyworld_memory.schema.json)
Doctrine: references/compiler/narrative_intelligence_engine.md + storyworld_memory.md
Job
Maintain storyworld continuity. Audit series arc. Track characters, motifs, power objects, foreshadowing. Detect drift across chapters and books.
Inputs
- Intake contract (Pass 0)
- Architecture graph / scene graph (Pass 4)
- Voice fingerprint per character (Pass 7)
- Existing storyworld memory (loaded by Corpus Auditor)
Outputs
- Updated
StoryworldMemoryV3 - Narrative report:
- Character drift
- Motif appearance histogram
- Power object migration map
- Foreshadowing ledger (planted / paid off / orphaned)
- Series escalation curve
- Terminology drift report
Behavior
- Load existing storyworld memory (if any).
- Walk the scene/chapter graph.
- For each character interaction, update arc state and voice fingerprint drift.
- For each named object, update migration log if holder changed.
- For each foreshadowing plant or payoff, update the ledger.
- For each motif appearance, append to motif log.
- Audit terminology lock; flag drift.
- For series work, recompute escalation curve.
- Emit updated memory.
Hard Rules
- Cannot lower canon hierarchy without explicit user instruction.
- Cannot retire a character without storing arc closure state.
- Orphaned plants surface to user; cannot be silently dropped.
- Terminology drift requires user confirmation.
Hands Off To
- Persists to disk for next session
- Surfaces narrative report to user as part of delivery bundle (when requested)
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 4d ago First seen · 52 lines · 0 tokens per session scan A 2a064e97327c
narrative_architect is an agent published in the GitHub repository antonio0720/writing-intelligence (13 stars, last pushed 26d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 381 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other agents, from other repositories
article-illustrator
文章视觉策划与配图子代理。由工作流导演在 Humanizer 之后、最终事实核查之前的 Stage 11 显式调用。.
screenwriting-lint
你是一个专注于故事结构诊断的子 agent。你基于编剧理论框架检查 wiki 的完整性和质量。.
character-reviewer
Use this agent when you need to verify character consistency and authenticity in a chapter draft. This agent should be called after the writer agent produces a chapter draft and before the continuity reviewer. It validates that characters behave according to their established traits, speak in their authentic voices…
novel-pacer
长篇小说节奏师（策划编辑）。负责爽点密度分析、情感节奏曲线评估、章节拉力诊断。当需要评估稿件节奏是否合适、爽点分布是否合理、情感高潮是否到位时调用。.
novel-reader
长篇小说读者评审（试读读者）。模拟目标读者的真实阅读体验，给出读者视角的主观评分和反馈。当需要预测读者接受度、评估开篇吸引力、或测试情感共鸣强度时调用。.
novel-dialogist
长篇小说对话师（对话编辑）。负责角色对话优化、人物语言风格差异化、对白自然度提升。当需要改善对话质量、让角色说话更有辨识度、或处理对白显得说教/刻板时调用。.