📄 CodeRabbit inference engine (AGENT.md)

Document agent architecture and design patterns in AGENTS.md

Update AGENTS.md API reference when adding new public NIF functions or Ecto features that users should know about

Files:

• AGENTS.md

📄 CodeRabbit inference engine (CLAUDE.md)

Use British/Australian English for all documentation files (except SQL keywords and compatibility requirements)

Files:

• AGENTS.md

📄 CodeRabbit inference engine (CLAUDE.md)

When ending a work session, complete the mandatory workflow: file issues for remaining work, run quality gates (tests, linters, builds), update issue status, commit changes with git commit, clean up stashes, and provide context for next session

Files:

• AGENTS.md
• lib/ecto_libsql/json.ex
• mix.exs

📄 CodeRabbit inference engine (CLAUDE.md)

**/*.{ex,exs}: Use British/Australian English for all Elixir code comments and documentation (except SQL keywords and compatibility requirements)
Always format Elixir code before committing using mix format --check-formatted

**/*.{ex,exs}: Use named parameters (:name, @name, or $name syntax) for clarity and maintainability in SQL queries with multiple parameters
Always handle transaction errors with proper error recovery using with clauses or case statements to ensure rollback on failure
Use prepared statements with automatic caching for repeated queries to achieve 10-15x performance improvement
Use batch operations (batch/batch_transactional) instead of individual inserts to reduce database roundtrips and improve performance
Use cursors and streams for large result sets instead of loading all rows into memory to prevent memory exhaustion
Always pattern match on error tuples from database operations rather than using destructuring assignment to avoid unhandled failures
Use savepoints for partial rollback within transactions to enable error recovery and batch import patterns
Always preload associations before accessing them to avoid N+1 query problems in Ecto queries
Use JSON helper functions from EctoLibSql.JSON module for safe JSON manipulation instead of string concatenation
Use vector operations for semantic search instead of LIKE pattern matching to provide more relevant search results based on embeddings
Use DBConnection.stream() with max_rows parameter for memory-efficient processing of large result sets

Files:

• lib/ecto_libsql/json.ex
• mix.exs

Learnt from: CR
Repo: ocean/ecto_libsql PR: 0
File: CLAUDE.md:0-0
Timestamp: 2026-01-06T06:36:01.571Z
Learning: Applies to AGENTS.md : Update AGENTS.md API reference when adding new public NIF functions or Ecto features that users should know about

Applied to files:

• AGENTS.md
• lib/ecto_libsql/json.ex
• mix.exs

Learnt from: CR
Repo: ocean/ecto_libsql PR: 0
File: AGENTS.md:0-0
Timestamp: 2026-01-06T06:36:18.238Z
Learning: Applies to **/*.{ex,exs} : Use JSON helper functions from EctoLibSql.JSON module for safe JSON manipulation instead of string concatenation

Applied to files:

• AGENTS.md
• lib/ecto_libsql/json.ex
• mix.exs

Learnt from: CR
Repo: ocean/ecto_libsql PR: 0
File: AGENTS.md:0-0
Timestamp: 2026-01-06T06:36:18.238Z
Learning: Applies to priv/repo/migrations/*.exs : Use RANDOM ROWID in migration table definitions to prevent ID enumeration attacks and improve security

Applied to files:

• mix.exs

Learnt from: CR
Repo: ocean/ecto_libsql PR: 0
File: AGENTS.md:0-0
Timestamp: 2026-01-06T06:36:18.238Z
Learning: Applies to lib/**/*schema*.ex : Use Ecto changesets with changeset/2 functions to validate and cast data before database operations

Applied to files:

• mix.exs