Research Radar

The description states:

• Research Radar is a Codex plugin that reads research data from Notion.
• It uses an MCP server, TypeScript, and GPT-5.6 to process structured evidence.
• It detects risk signals such as repeated failures, missing steps, or blocked work.
• It outputs a Rescue Plan with:
  • Transparent risk level.
  • Source-linked factual evidence.
  • Clearly labeled inferences.
  • At most three prioritized actions.
  • Preview of proposed Notion changes.

The plugin is described as installable and supports both live (with credentials) and demo modes.

• Demo mode uses anonymous fixture data and produces Markdown files.
• Live mode creates Notion pages and annotates task pages, with confirmation required before writing.

Inference The product appears to be a research-focused tool that attempts to automate decision-making in experimental workflows by linking Notion records to AI-generated action plans. It is not a general-purpose assistant but a narrow, domain-specific plugin for lab researchers.

The description states:

• The project was built to help experimental researchers who lack reliable ways to turn scattered notes into actionable decisions.
• It aims to prevent repetition of failed experiments, loss of evidence trails, or late discovery of blocked milestones.
• The author emphasizes that the most valuable research assistant is one that shows why an action is recommended, links to evidence, and gives control at the point of change.

Inference The positioning is narrow and targeted — it does not claim to be a general-purpose research tool or Notion assistant. It positions itself as a risk mitigation plugin for experimental workflows, with emphasis on transparency, provenance, and user control.

The description states:

• The target customer is experimental graduate researchers.
• The product is built to help them avoid repeating failed experiments or losing track of deadlines.

Inference The ICP appears to be a specific segment of researchers — likely in STEM fields, working with lab protocols and Notion for task management. No evidence of broader targeting or segmentation beyond this group.

The description states:

• The plugin is installable, but no pricing model or monetization strategy is described.
• It supports both demo mode (no credentials) and live mode (with Notion integration).
• No mention of subscriptions, usage fees, or paid features.

Inference There is no evidence of a business model. The project is described as a hackathon MVP with no indication of monetization plans or pricing.

The description states:

• Built using Codex, GPT-5.6, and TypeScript stdio MCP server.
• Uses an internal Notion integration token to read data.
• Implements a structured evidence contract with deterministic checks for risk signals.
• GPT-5.6 is used to explain implications, not summarize broadly.
• Includes unit and protocol-level tests, and automated client/server components.

Inference The technical stack suggests a lightweight, developer-oriented tool built for integration with Notion and AI. It is described as idempotent and auditable, with explicit confirmation steps before writing.

The description states:

• This is a hackathon submission (OpenAI 2026).
• No evidence of revenue, customers, or adoption.
• The demo works with anonymous fixture data, not real user data.
• The author notes that future versions may include:
  • Public Notion OAuth
  • Configurable lab templates
  • Scheduled weekly reviews
  • Experiment-instrument imports

Inference No traction or maturity signals are evident. The project is described as a prototype, and no evidence of real-world usage or product-market fit exists.

The description states:

• No mention of competitors.
• The author does not reference existing tools for experimental research or Notion-based workflows.

Inference There is no evidence of competitive analysis or awareness of existing solutions. The project appears to be self-contained and unanchored in a known marketplace.

The description states:

• The demo mode works without credentials, but the live version requires Notion integration.
• Risk signals are deterministic checks, but model interpretation is used for explanations.
• The system is described as idempotent and auditable, but no evidence of safety or error handling beyond tests.

1. What is the actual research workflow you’re targeting? Is there any user feedback or pilot testing?
2. How does the plugin handle data privacy and access control in Notion?
3. Are there plans to integrate with other tools beyond Notion?
4. What are the key assumptions about how researchers will interact with the Rescue Plan?
5. Has the team considered how to scale the deterministic checks for broader use cases?