Labels

Agent runtime ownership, architecture boundaries, and typed events

Terminal UI and interactive agent loop

Legacy protocol boundaries, migration, and deprecation

Canonical conversation history, compaction, and context efficiency

Token usage, pricing, accounting, and spend reporting

Git tools and repository operations

Installer, packaging, distribution, and development setup

Background learning, provenance, evaluation, and promotion

MCP discovery, capabilities, calls, and protocol behavior

Memory retrieval, storage, scope, and turn snapshots

SQLite state, schema migration, event history, and recovery

LLM providers, models, retries, and streaming clients

Model routing, reasoning budgets, complexity, and Jev advice

Capabilities, approvals, scope boundaries, and OS sandboxing

Modular workflow guidance, skill versions, and turn snapshots

Sub-agent profiles, execution, evidence, and learning

Task planning, execution, recovery, and evidence

Test dependencies, fixtures, and acceptance coverage

Workspace file search and edit tools

FastAPI Web surface and browser chat UI

Command-line and terminal UI

Something isn't working

Improvements or additions to documentation

This issue or pull request already exists

New feature or request

Good for newcomers

Extra attention is needed

This doesn't seem right

Critical Issue