Commit Graph

3 Commits

Author SHA1 Message Date
vikingowl 43ea2e562d feat(engine): two-stage tool routing for small local arms
Plan A from docs/superpowers/plans/2026-05-19-post-slm-unlock.md.

Small local SLMs (<=16k context) waste ~1500 tokens per turn on the
full tool catalogue. Two-stage routing replaces round-1 tools with a
single synthetic select_category schema; round-2+ sends only the
selected category's real tool schemas plus select_category for
re-selection.

- internal/tool/category.go: Category type, optional Categorized
  interface, CategoryOf() with meta fallback. fs.read/fs.ls -> read,
  fs.write/fs.edit -> write, fs.glob/fs.grep -> search, bash -> exec.
- internal/engine/twostage.go: synthetic select_category tool,
  intercept helper, per-turn selectedCategory state under e.mu.
- Engine round 1 forces ToolChoiceRequired so SLMs don't fall back to
  prose. State resets at the top and end of every runLoop.
- Activates automatically on a forced local arm with ContextWindow
  <=16384, or via [router].force_two_stage TOML key.
- Integration test drives a 3-round trip and asserts: round 1 emits
  exactly one schema (synthetic) with ToolChoiceRequired, round 2
  contains only write-category schemas + select_category, real
  fs.write executes. Invalid-category fallback round-trips back to
  round-1 mode.
2026-05-19 20:53:21 +02:00
vikingowl 0b1392cf6b feat(pty): Phase 2 — interactive shell and bash interactive detection
- /shell [cmd]: launch user's $SHELL via tea.ExecProcess (PTY handoff)
  hands terminal to the shell and restores TUI on exit.
  /shell <cmd> runs that command in the shell directly.
  Detects $SHELL > $COMSPEC > /bin/sh|powershell.exe in order.

- bash tool: detect interactive commands before execution
  Prefix-interactive: sudo, ssh, passwd, vim/vi/nano, less/more,
  htop/top, mysql/psql, ftp/sftp, git push.
  Exact-interactive (REPL): python3/python/node/irb/iex/ghci/julia.
  Returns a tool result with interactive=true metadata and a hint to
  use /shell instead of hanging or erroring.

- completions: add /shell to builtin command list
- help: document /shell [cmd]
2026-05-07 15:52:56 +02:00
vikingowl 69f5dba091 feat: complete M1 — core engine with Mistral provider
Mistral provider adapter with streaming, tool calls (single-chunk
pattern), stop reason inference, model listing, capabilities, and
JSON output support.

Tool system: bash (7 security checks, shell alias harvesting for
bash/zsh/fish), file ops (read, write, edit, glob, grep, ls).
Alias harvesting collects 300+ aliases from user's shell config.

Engine agentic loop: stream → tool execution → re-query → until
done. Tool gating on model capabilities. Max turns safety limit.

CLI pipe mode: echo "prompt" | gnoma streams response to stdout.
Flags: --provider, --model, --system, --api-key, --max-turns,
--verbose, --version.

Provider interface expanded: Models(), DefaultModel(), Capabilities
(ToolUse, JSONOutput, Vision, Thinking, ContextWindow, MaxOutput),
ResponseFormat with JSON schema support.

Live verified: text streaming + tool calling with devstral-small.
117 tests across 8 packages, 10MB binary.
2026-04-03 12:01:55 +02:00