Repo Encoder

Use research-and-incorporate instead when:

• Researching an external topic, not encoding a codebase
• You need analysis of a single concept, not full repository encoding

Quick Start

/repo-encoder
/repo-encoder ./my-project
"encode this repository"
"populate brain with this codebase"

Prerequisites

1. Serena plugin: claude plugins list | grep serena
2. Brain MCP: Test with mcp__plugin____brain__search({ "query": "test" })
3. If missing, set up Brain MCP first

Process

Phase 0: Discovery

Assess project size, complexity, and structure. Produce a structure map.

Phase 1: Foundation

Create 5-10 project overview Brain memory notes covering purpose, tech stack, and entry points.

Phase 1B: Dependencies

Create 1-3 dependency notes documenting external libraries and internal references.

Phase 2: Symbols

Use Serena find_symbol and find_referencing_symbols to produce 10-15 architecture notes.

Phase 2B: Entities

Create component Brain memory notes with relations. Deduplicate before creating.

Phase 3: Patterns

Document 8-12 recurring code patterns, conventions, and idioms.

Phase 4: Features

Create 1-2 notes per critical feature describing behavior and implementation.

Phase 5: Decisions

Record design decisions with rationale and alternatives considered.

Phase 6: Artifacts

Store code artifacts (configs, schemas, key files) as Brain memory notes.

Phase 6B: Symbol Index

Create a symbol index note for navigation.

Phase 7: Documents

Produce long-form documentation summarizing the codebase.

Phase 7B: Architecture

Create an architecture reference note linking all prior phases.

See references/phases.md for full phase details.

Execution Order

0 -> 1 -> 1B -> 2 -> 2B -> 3 -> 4 -> 5 -> 6 -> 6B -> 7 -> 7B

Guidelines:

• Execute phases in order
• Use Serena's find_symbol and find_referencing_symbols
• Deduplicate notes before creating
• Link notes via relations bidirectionally
• Create summary notes for large documents

Memory Targets

Quality Principles

Validation

After encoding, verify all outputs meet quality standards:

• Test search: "How do I add a new API endpoint?"
• Test dependency query: "What dependencies does this project use?"
• List notes by folder
• Verify note relations
• Check Symbol Index note exists
• Check Architecture Reference note exists
• All target components have Brain memory notes
• Notes linked via relations
• No duplicate notes in knowledge graph

See references/validation.md for test commands.

Anti-Patterns

References

Related Skills

• /code-symbols - Serena symbol analysis
• /code-architecture - Architectural analysis with Brain memory