Research Status

Research Status Overview

Generate a comprehensive status snapshot of an research project.

Context

At any point during a research project, stakeholders need to understand: where the research stands, what has been learned, what the active fronts are, and how much budget remains. This skill reads all project artifacts and produces a structured overview.

This is especially useful for quarterly review preparation, onboarding new team members, and periodic self-assessment by the researcher.

Procedure

Step 1 — Locate the research project

Ask the user to identify the research project (by ID, name, or directory path). Read the brief (brief.md) to confirm the correct project.

Step 2 — Gather data

Read the following files:

1. brief.md — research question, success criteria, scope, status,

constraints, implementation plan.

1. hypotheses/graph.md — the Mermaid diagram and catalog table.
2. All hypothesis cards (hypotheses/H-*.md) — to get detailed status

and results.

1. prior-art.md — to count cataloged sources.

Step 3 — Compute metrics

Calculate:

• Total hypotheses: count of all H-NNN.md files.
• By status: how many are open, in-progress, confirmed, refuted,

cancelled.

• Confirmation rate: confirmed / (confirmed + refuted), excluding

cancelled and open.

• Graph depth: longest path from a root hypothesis to a leaf.
• Graph breadth: maximum number of active (non-terminal) branches.
• Active front: hypotheses that are open or in-progress.
• Prior art entries: count of entries in prior-art.md.

If the brief contains time budget information (start date, expected end date), calculate:

• Elapsed time: from start to today.
• Remaining time: from today to expected end.
• Time consumption: elapsed / total as a percentage.

Step 4 — Identify the current front line

The "front line" is the set of hypotheses at the leading edge of the research — those that are currently open or in-progress, and represent the active areas of investigation. For each front-line hypothesis:

• ID and statement.
• Timebox status (how much time remains).
• Parent chain: trace back to root to show the reasoning path.

Step 5 — Summarize key findings

From confirmed and refuted hypotheses, extract the most important findings — things that answer (partially or fully) the research question, or that significantly shaped the research direction.

List the top 3–5 findings with one sentence each.

Step 6 — Identify risks and concerns

Flag potential issues:

• Timebox overruns: hypotheses where the timebox is close to or past

expiry without a result.

• Stalled branches: hypotheses in in-progress with no recent updates.
• Budget pressure: if overall time consumption exceeds hypothesis

completion rate, flag convergence risk.

• Scope drift: if the active hypotheses have diverged significantly

from the original research question.

Step 7 — Generate the overview

Present the status in a structured format:

# Research Status: [R-NNN] Title

## Summary
- Status: Active / Completed / Paused
- Research question: ...
- Time: X of Y days elapsed (Z%)
- Hypotheses: A total (B confirmed, C refuted, D cancelled, E open/in-progress)

## Current Front Line
- H-NNN: [statement] — [timebox status]
- H-NNN: [statement] — [timebox status]

## Key Findings So Far
1. [finding]
2. [finding]
3. [finding]

## Hypothesis Graph
[Mermaid diagram from graph.md]

## Risks and Concerns
- [risk]
- [risk]

## Prior Art
- X sources cataloged (Y high relevance, Z medium, W low)

## Implementation Plan Status
[Current state of the implementation plan from brief.md]

Step 8 — Suggest next actions

Based on the overview, suggest what the researcher should focus on next:

• Which hypothesis to experiment on next (if multiple are open).
• Whether any branch should be killed due to resource pressure.
• Whether synthesis should begin (if the research question is answered

or budget is nearly exhausted).

• Whether additional prior art research would be valuable.

Quarterly Review Mode

When invoked with a quarterly review context, additionally prepare:

• Changes since last review: compare current state with the state at

the start of the quarter (or last review date if provided).

• Demo-ready prototypes: list prototypes that can be demonstrated.
• Next quarter plan: proposed focus areas and hypotheses for the

next period.

Relation to Other Skills

• Can be invoked at any time during the research lifecycle.
• Feeds into research-decision (informs graph-level decisions).
• Provides the starting data for research-synthesis.
• Useful before invoking research-prior-art (identifies knowledge gaps).

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: v0lka
• Source: v0lka/skills
• License: MIT

Install and usage instructions live in the source repository linked above.