When is the best time to use Research Capture?

You should trigger it after completing a technical spike, making a significant design choice, discovering a non-obvious solution, or at the end of a development session when prompted.

What is the difference between a finding and a decision?

A decision records a specific architectural or design choice along with its trade-offs and rationale. A finding documents a discovery or insight gained during development, such as a subtle bug or an undocumented API behavior.

Can I search through my captured research later?

Absolutely. You can use the /popkit:research search command to find entries by keyword or semantic similarity, helping you recall why specific decisions were made months ago.

Does this skill work offline?

Yes. Research entries are stored locally in your project's .claude/research folder. While semantic search features require a cloud API key for embeddings, basic keyword searching and entry creation work entirely offline.

Research & Decision Capture

Name: Research & Decision Capture
Author: jrc1883

byjrc1883

Productivity & Workflow

Captures and indexes architectural decisions, research findings, and technical learnings for long-term project knowledge.

About

Research & Decision Capture serves as a persistent memory for your development process by documenting significant technical discoveries and architectural choices as they happen. It transforms fleeting insights into structured, searchable records—complete with context, rationale, and alternatives considered—ensuring that the 'why' behind your code is never lost. By utilizing semantic search and cloud embedding integration, the skill proactively surfaces relevant past research during current development tasks, helping to prevent redundant investigations and highlighting potential conflicts with previous design choices.

Key Features

0 GitHub stars
Structured capture of architectural decisions, spikes, findings, and learnings
Automatic context gathering from Git branches, commits, and project metadata
Semantic retrieval using cloud-synced embeddings for intelligent searching
Automated surfacing of relevant historical research during new development tasks
End-of-session integration to ensure insights are captured while context is fresh

Use Cases

Recording complex findings about race conditions or third-party API quirks to prevent future regressions
Onboarding new developers by providing a searchable history of why the project was built a certain way
Documenting the rationale and rejected alternatives for choosing a specific database or framework

About

Key Features

0 GitHub stars
Structured capture of architectural decisions, spikes, findings, and learnings
Automatic context gathering from Git branches, commits, and project metadata
Semantic retrieval using cloud-synced embeddings for intelligent searching
Automated surfacing of relevant historical research during new development tasks
End-of-session integration to ensure insights are captured while context is fresh

Use Cases

Recording complex findings about race conditions or third-party API quirks to prevent future regressions
Onboarding new developers by providing a searchable history of why the project was built a certain way
Documenting the rationale and rejected alternatives for choosing a specific database or framework