Commit Graph

9 Commits

Author SHA1 Message Date
vikingowl ebfbefc73d feat: configurable max_turns for elfs — LLM sets via agent tool param 2026-04-03 19:37:17 +02:00
vikingowl 086a4622fd fix: elf progress — proper last-2-lines tracking, 70 char truncation 2026-04-03 19:30:18 +02:00
vikingowl c01069164e feat: live elf progress in TUI
- Elf tool calls show as 🦉 [elf] <prompt> (not ⚙ [agent])
- Live 2-line progress beneath the elf label showing what the
  elf is currently outputting (grey, auto-updated)
- Agent tool forwards elf streaming events via progress channel
- Progress cleared on turn completion
- elfProgressCh wired from agent tool → TUI
2026-04-03 19:25:43 +02:00
vikingowl 13db7521b1 feat: M7 Elfs — sub-agents with router-integrated spawning
internal/elf/:
- BackgroundElf: runs on own goroutine with independent engine,
  history, and provider. No shared mutable state.
- Manager: spawns elfs via router.Select() (picks best arm per
  task type), tracks lifecycle, WaitAll(), CancelAll(), Cleanup().

internal/tool/agent/:
- Agent tool: LLM can call 'agent' to spawn sub-agents.
  Supports task_type hint for routing, wait/background mode.
  5-minute timeout, context cancellation propagated.

Concurrent tool execution:
- Read-only tools (fs.read, fs.grep, fs.glob, etc.) execute in
  parallel via goroutines.
- Write tools (bash, fs.write, fs.edit) execute sequentially.
- Partition by tool.IsReadOnly().

TUI: /elf command explains how to use sub-agents.
5 elf tests. Exit criteria: parent spawns 3 background elfs on
different providers, collects and synthesizes results.
2026-04-03 19:16:46 +02:00
vikingowl 60883521c7 feat: auto permission mode, edit diffs, truncated tool output
- Default permission mode changed to 'auto' (read-only auto-allows,
  writes prompt)
- fs.edit now shows diff-style output: line numbers, context ±3 lines,
  + for added (green), - for removed (red)
- Tool output truncated to 10 lines in TUI with "+N lines (Ctrl+O
  to expand)" indicator
- Mistral SDK bumped to v1.3.0
2026-04-03 18:57:13 +02:00
vikingowl 46505a1f71 feat: complete 7/7 bash security checks
Added:
- Standalone semicolon check: blocks ; outside quotes (use && instead)
- Sensitive redirection check: blocks > to /etc/passwd, .bashrc,
  .ssh/authorized_keys, .env, etc.

Now all 7 security checks are active:
1. Incomplete commands, 2. Control characters, 3. Newline injection,
4. Command substitution, 5. Dangerous variables, 6. Semicolons,
7. Sensitive redirections
2026-04-03 17:56:01 +02:00
vikingowl 6cfe35620d feat: compact system inventory with queryable system_info tool
System prompt gets a one-line summary (~200 chars): OS, CPU, RAM,
GPU, top runtimes, package count, PATH command count.

Full details available on demand via system_info tool with sections:
runtimes, packages, tools, hardware, all. LLM calls the tool when
it needs specifics — saves thousands of tokens per request.

Hardware detection: CPU model, core count, total RAM, GPU via lspci.
Package manager: pacman/apt/dnf/brew with dev package filtering.
PATH scan: 5541 executables. Runtime probing: 22 detected.
2026-04-03 14:50:33 +02:00
vikingowl 8e5ddb20cb feat: hybrid system inventory — dynamic PATH scan + runtime probing
No hardcoded tool lists. Scans all $PATH directories for executables
(5541 on this system), then probes known runtime patterns for version
info (23 detected: Go, Python, Node, Rust, Ruby, Perl, Java, Dart,
Deno, Bun, Lua, LuaJIT, Guile, GCC, Clang, NASM + package managers).

System prompt includes: OS, shell, runtime versions, and notable
tools (git, docker, kubectl, fzf, rg, etc.) from the full PATH scan.
Total executable count reported so the LLM knows the full scope.

Milestones updated: M6 fixed context prefix, M12 multimodality.
2026-04-03 14:36:22 +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