An intelligent preprocessing skill designed to parse arbitrary script inputs, classify their development level, and extract structural metadata to guide downstream content generation.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install editor-classify-and-parse
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install editor-classify-and-parse using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is editor-classify-and-parse?
The editor-classify-and-parse skill is an essential pipeline component within the ecosystem of Openclaw Skills. It functions as a read-only analyzer that evaluates any incoming content or script format, deterministically mapping it to a structural maturity level from Level -1 to Level 4. By establishing the structural baseline of the text, this skill acts as a crucial triage system for AI agents, ensuring subsequent generative processes map accurately to the user's intent and level of detail.
Beyond simple classification, it breaks down rich text elements from higher-level inputs (Level 2 and above) into machine-readable JSON schemas. It extracts foundational storytelling assets including scene boundaries, character registers, dialogue weights, and action densities. This structured intelligence empowers multi-agent configurations to smoothly collaborate with clear operational boundaries.
editor-classify-and-parse Use Cases
- Script Triage and Routing: Automating the classification of incoming creative documents to determine whether they require plot structure validation, dialogue expansion, or storyboarding.
- Structural Quality Assurance: Detecting script anomalies, such as an insufficient density of physical action lines relative to dialogue cues, before sending files to production workflows.
- Asset and Metadata Extraction: Instantly compiling character lists, initial appearance trackers, and props catalogs from raw, unstructured screenplays or teleplays.
How editor-classify-and-parse Works
- Availability Verification: The skill checks for the existence of input data; if empty, it triggers an absolute Level 0 status.
- Premise Evaluation: It verifies the core story elements; if crucial dramatic components like Want, Need, or Stakes are absent or lack tension, it flags the input as Level -1.
- Scene and Dialogue Boundaries Detection: The script scans text for specialized scene headers and character dialogue blocks to differentiate an open narrative outline (Level 1) from structured dialogue layouts (Level 2).
- Cinematic Shot & Directorial Intent Check: It analyzes physical descriptors for camera movements or visual staging markers to identify rough storyboards (Level 3) or completely realized production drafts (Level 4).
- Structural Mapping and Extraction: For scripts ranked Level 2 or higher, the skill isolates metadata components, performing statistical counts on speech lines, character appearances, and embedded scene items.
- JSON Output Delivery: The calculated dimensions, confidence score, and potential structural risk flags are rendered into a consolidated telemetry payload.
editor-classify-and-parse Setup
To deploy this module within your local runtime ecosystem, pull down the package from your repository and integrate it into your workspace manifest:
openclaw install skill-editor-classify-and-parse
Configure the tool settings inside your master workflow protocol configuration file to activate Phase 2 integration.
editor-classify-and-parse Data Schema & Taxonomy
The skill translates raw text artifacts into structured telemetry data. Below is the system taxonomy applied to parsed outputs:
| Attribute | Type | Description |
|---|---|---|
detected_level |
Integer | Calculated hierarchy indicator from -1 (invalid premise) to 4 (highly detailed script) |
level_confidence |
Float | Probability metric (0.0 - 1.0) of the level classification accuracy |
level_evidence |
String | Plain-text summary outlining the volume of discovered structural elements supporting the verdict |
structure.scene_count |
Integer | Total count of unique scene headers parsed |
structure.character_list |
Array | Exhaustive list of unique character entities recognized within text cues |
structure.first_appearances |
Object | Key-value mapping tracking exactly which scene each character debuts in |
structure.dialogue_stats |
Object | Character-by-character ledger documenting the breakdown of speech line counts |
structure.action_density |
Float | Calculated ratio comparing physical description lines against dialogue lines |
structure.props |
Array | Items, tools, costumes, or special elements identified in the text |
risk_flags |
Array | Operational alerts generated if script metrics violate established design thresholds |
editor-classify-and-parse Advanced Features
- Multi-Agent Orchestration Hooks: Synchronizes directly with backend modules like editor-story-architect and editor-consistency-checker to build integrated character rosters and script compliance sheets.
- Automated Validation Guardrails: Monitors formatting boundaries, tracking an 'Action Density Threshold' to ensure scripts adhere to standard narrative balances, such as requiring at least 1 action line per every 3 lines of dialogue.
- Escalation Exception Handling: Automatically shifts into a
help_requestedexception mode if classification confidence falls below 0.7, avoiding misinterpretations in ambiguous user drafts. - Strict High-Tier Input Protection: Enforces immutable preservation rules on complex Level 3/4 scripts to shield raw dialogue and staging from unauthorized generative modifications during downstream tasks.
SKILL.md
Loading
Related Openclaw Skills
Designing TUIs
An advanced AI-driven design system to decompose, design, and programmatically generate stunning Terminal User Interfaces (TUIs) from screenshots, sketches, or text descriptions.
RUS-CHN05 Bone Age Analyzer
An AI-powered diagnostic helper that evaluates left-wrist X-ray images using the Chinese RUS-CHN05 standard to deliver precise skeletal maturity scores and genetic height predictions.
Byted Kickart Viral Replicator
An advanced short video replication and cloning skill for Volcengine Kickart, allowing creators to automatically reproduce viral video structures, scripts, and visual styles.
LuckyBall (Shuangseqiu Lucky Number Predictor)
LuckyBall v3.0 merges Chinese Bazi metaphysics and Pointwise Random Forest models with AR(1) mean-reversion to analyze and forecast Shuangseqiu lottery numbers.
trade-xyz-with-cai
Execute HIP-3 market orders on trade.xyz utilizing CAI tools with your existing Hyperliquid custodial stack.
SciSurvey
SciSurvey is an automated systematic review and academic survey generator that leverages Sciverse MCP to automate the discovery, extraction, mapping, and synthesis of research papers.
Featured*
