The whole plan shipped. Now we write the next one.

Every item on the original roadmap is now in a release. The next wave below is being scoped; say what you want first in Discussions.

• All 168 roadmap items checked off, capped by the v1.22.0 CLI-first round
• 86 commands, fully documented and generated straight from --help

• Autonomous long-horizon projects spanning days
• Fine-tuned repo-specific models
• On-device small-model routing for trivial tasks
• Agent federation across organizations
• AI release manager: cut, verify, and publish releases end to end

• bolt ask, --json envelopes, --porcelain, stdin context, and a stable exit-code catalog
• Run chaining (--emit context), bolt exec playbooks, bolt batch, and --output-schema
• Plan-only CI gate, bolt diff-gate, and the official bolt-action GitHub Action
• Sessions from the shell: ls filters, fuzzy resume, tags, fork, grep, prune, export --md/--jsonl
• Session handoff between machines with bolt push and bolt pull
• Named profiles, config doctor/get/set/unset/diff, BOLT_* env overrides, init presets, aliases
• Sub-50ms startup, bolt daemon, bolt warm, prompt-cache reuse, --offline
• bolt doctor, bolt stats, logs --follow, crash reports, release channels, clean uninstall
• Remote exec over SSH, bolt mux split panes, serve --api, plugin subcommands, voice input

• bolt learn: repo convention learning saved into project memory
• Compounding memory: every session teaches the next one
• bolt memory why: ask why the agent believes something and where it learned it
• Cross-session recall: "do it like we did in that auth refactor last month"
• Memory decay, conflict resolution, provenance, search, and per-directory scopes
• Team memory: opt-in shared project memory across teammates

• bolt refactor: test-aware loops that change, run, verify, repeat until green
• Confidence scoring: the agent tells you when it is guessing
• bolt review --staged / --branch: one-shot AI diff review with exit codes
• bolt commit: commit messages you don't have to rewrite
• Self-review pass before any diff is handed to you
• Mutation testing, property-based test generation, and regression guards

• bolt watch: tests rerun on every save and failures fix themselves
• bolt cron: scheduled agent chores (dependency bumps, changelog drafts, issue triage)
• bolt bisect: find the culprit commit, show blame, and propose the fix
• bolt flaky: flaky test detection and quarantining
• Background job queue: run --background plus bolt jobs list/tail/kill

• Guardrail agent that vetoes risky commands before they run
• Budget guards: --max-cost and --max-tokens on any run
• bolt undo: one-command rollback when an experiment goes sideways
• Named checkpoints: save points you can rewind the repo and the conversation to
• Secrets firewall, protected paths, dry-run mode, and blast-radius estimates

• Multi-agent pipelines: one agent plans, one codes, one reviews
• Automatic agent selection based on prompt intent
• bolt figma: design-to-code from a Figma node
• bolt map: codebase visualization maps drawn by the agent
• bolt pair: one live session shared across two terminals
• Session replays: export --html writes a shareable step-through timeline

• Best-of-N runs: the same task fanned out to several models in parallel, keep the winner
• Push-to-talk voice input in the TUI
• Agent evaluation benchmark harness (bolt eval)
• Project memory module: automatic capture with save and recall tools
• bolt memory: view and edit what the agent remembers per project (/memory in the TUI)
• Per-session memory controls
• /plan mode, /todos task list, and /btw side questions that don't interrupt the task
• bolt logs with tail and follow
• .boltignore-aware agent file search
• macOS desktop app with a one-line terminal installer (no Gatekeeper detour)