← Back to Terminal

Sovereign Querying: How to Research Literary Agents with AI and Automate MSWL Match Analysis

The legacy publishing pipeline is bottlenecked by subjective screening. For the serious author, the query process is typically an exercise in administrative degradation—manually sorting through hundreds of disparate web pages to find a fraction of alignment. Under the Maha Strategies paradigm, we treat this search not as a creative lottery, but as a precise database querying problem.

This technical documentation outlines how to deploy the publish-analyze_mswl protocol, establishing a sovereign standard for how to research literary agents with AI and demonstrating why agentic systems represent the best AI tools for analyzing agent MSWL (Manuscript Wish List) data.

1. The Legacy Friction: The MSWL Inefficiency

The manual querying process is a systemic failure of author resource allocation. Traditional methods require authors to spend dozens of hours scouring personal blogs, Twitter feeds, and the central MSWL database to discern what an agent is actively seeking. This approach yields a terrible return on a writer's time for three reasons:

  • Semantic Drift: An agent's stated preferences ("high-concept speculative fiction with strong character voices") are highly subjective and drift over time.
  • Cognitive Asymmetry: Human authors are forced to manually cross-reference their 100,000-word manuscript's deep thematic layers against vague, 200-word agent descriptions.
  • Zero Optimization: The manual process produces high-latency, generic query hooks that fail to capture the precise intersections between the manuscript's architectural core and the agent's active sub-genre desires.

2. The Agentic Solution: Programmable Alignment

The publish-analyze_mswl tool bypasses this manual triage entirely. Instead of human-matching, it treats the agent's Manuscript Wish List and the author's intellectual property as two structured datasets.

By executing this tool, the system programmatically scrapes and ingests the raw MSWL text. It then executes a semantic cross-reference against the core components of the book's architecture (specifically, the structural chapters, central thesis, and ideological paradigms). Rather than relying on superficial keyword matching, the tool evaluates deep conceptual resonance—determining whether an agent's wish list aligns with core thematic pillars such as Metabolic Colonialism, Attentional Captivity, or the Nurturing Warrior archetype. The result is a quantitative alignment score and a targeted query hook generated in milliseconds.

3. Technical Breakdown of the publish-analyze_mswl Schema

The tool is defined as an idempotent, read-only system call within the Maha Gateway registry. Below is the precise input and output schema configuration:

// Tool Schema Definition

Name: publish-analyze_mswl

Properties (Input):

  • agentName (string, required): The literal name of the target literary agent.
  • mswlText (string, required): The raw, scraped text containing the agent's stated manuscript interests, sub-genres, and aesthetic preferences.

Annotations: { "readOnlyHint": true, "idempotentHint": true }

Pipeline Data Processing

When the tool is invoked, the gateway processes the payload through the following pipeline:

  1. Ingestion & Compilation: The tool reads the local, public ecosystem files—specifically the core architecture document (maha-framework.md) and the primary book proposal (book-proposal.md).
  2. Context Assembly: It constructs a high-fidelity prompt combining the agent's raw MSWL text, the complete book proposal, and the theoretical parameters defined in the framework document.
  3. Algorithmic Matching: The compiled payload is dispatched to our localized guardianModel (powered by the Gemini engine). The model evaluates semantic overlap on a multi-dimensional matrix.
  4. Structured Output Generation: The server returns a structured JSON object containing a strict match percentage (0-100), matching themes, and a tailored, high-converting two-sentence query hook.

By deploying this protocol, authors convert querying from a desperate, high-friction pitch into a targeted B2B transaction.