sina-research-intake

Submits research findings to the Sina Agent Hub intake-brainstorm-evaluate-promote pipeline.

Updated Jun 20, 2026
One-click install
npx skills add https://github.com/Noetfield-Systems/SourceA --skill sina-research-intake-noetfield-systems
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: sina-research-intake
Source: https://github.com/Noetfield-Systems/SourceA/tree/main/.cursor/skills/sina-research-intake
Command: npx skills add https://github.com/Noetfield-Systems/SourceA --skill sina-research-intake-noetfield-systems

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Research findings, external document reviews, and cross-agent insights often go straight into code changes without evaluation, causing unvetted structural decisions. This Skill routes every finding through a governed pipeline before it can affect implementation. ## Core Features & Use Cases - Structured Research Submission: Submit findings via the Agent Hub UI at http://127.0.0.1:13020/?tab=agent-loop or by POSTing to /api/agent-research with title, body, source agent, and target agents. - Mandatory Evaluation Pipeline: Enforces the INTAKE → BRAINSTORM → EVALUATE → PROMOTE sequence, requiring at least one brainstorm note and a composite score of 65 or higher before promotion. - Governance Rules: Classifies external pastes as EXTERNAL_CRITIC (L1) and blocks workstream changes based on research alone (L3). - Use Case: After pasting a ChatGPT analysis that suggests reordering build steps, submit it through the hub so it is brainstormed and scored before any code changes are made. ## Quick Start Submit my completed research findings to the Sina Agent Hub research pipeline for brainstorming and evaluation.

Frequently Asked Questions about sina-research-intake

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I submit research to the Sina Agent Hub pipeline?

Open the hub at http://127.0.0.1:13020/?tab=agent-loop and use the Research pipeline form, or POST to /api/agent-research with action set to submit, a title of 8+ characters, a body of 40+ characters, source agent, and target agents.

What is the intake brainstorm evaluate promote pipeline?

It is a mandatory four-stage review flow for research with structural impact. After intake, you add at least one brainstorm note, run the evaluate action, and promote only if the composite score is 65 or higher.

When should research go through the hub instead of straight to code?

Any finished research, comparison, external paste review, or cross-agent insight that could affect build order or structure must go through the pipeline. Governance rule L3 forbids changing the workstream step from research alone.

Can I promote research without running the evaluate step?

No. The pipeline requires at least one brainstorm note followed by the evaluate action. Promotion is only allowed when the composite evaluation score reaches 65 or above.

How are external pastes like ChatGPT output classified?

External pastes are classified as INPUT CLASS: EXTERNAL_CRITIC at level L1 under the governance rules. They must be submitted through the research pipeline rather than applied directly to code.