marfrit
955bd82efb
Closes the last #13 gap — Norris broker call + is_destructive LLM
second-opinion probe were the two egress points NOT covered by the
scrub-at-egress design in commit d852aca.
Approach: option (b) per #52's fix sketch — callback-via-helpers/opts.
safety.lua does NOT gain a require("secrets") dependency (acceptance
criteria 3); integration is purely through the convention the rest
of the helpers table already uses.
safety.lua changes:
- llm_probe gains an opts table. When opts.scrub_msgs is set, the
{system, user(cmd)} message pair is scrubbed before broker.chat.
When opts.rehydrate is set, the YES/NO reply is rehydrated before
parsing (defensive — the verdict shouldn't carry placeholders but
rehydration is a safe no-op if it doesn't).
- llm_second_opinion threads opts through to llm_probe.
- M.is_destructive(cmd, cfg, opts) — opts optional; nil-opts is
backwards-compatible (no scrub, original behavior).
- M.norris_step:
* outbound broker.chat_stream message scrubbed via
helpers.scrub_msgs(ctx:to_messages(), model_cfg) when provided.
* on_delta wrapped with helpers.streaming_rehydrator():push /
:flush so the user sees rehydrated text AND text_parts
accumulates rehydrated chunks (parity with ask_ai in repl.lua).
* both M.is_destructive call sites (tool_call probe + CMD: probe)
now pass probe_opts = {scrub_msgs, rehydrate} when the
helpers carry them.
repl.lua changes:
- Norris helpers table gains scrub_msgs / rehydrate /
streaming_rehydrator closures, all nil-safe (return identity /
nil when secrets_session is nil).
- :safety check meta passes probe_opts to is_destructive when
secrets_session is configured. Without secrets, behavior unchanged.
Unit-test verified end-to-end:
- Stubbed broker.chat captures the messages it receives.
- Without opts: probe SEES `ghp_realsecretvalue_...` (control).
- With opts: probe sees `$AISH_SECRET_NNN` (correct scrub).
Regression: test_safety 87/87, test_router_model 31/31, repl loads.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
aish
aish — AI-augmented conversational shell.
A single REPL that interleaves shell command execution and language-model conversation, backed by a llama.cpp HTTP broker. Implementation is LuaJIT 2.x with FFI bindings to libcurl, GNU readline, and libc — no C extensions, no build step, one source tree.
Why
Three flows that currently live in three windows fold into one:
- "Run this command and show me the output" — fast feedback loop, no copy-paste between terminal and chat.
- "Explain or write code based on the output we just looked at" — exec output is automatically injected into the model's context.
- "Plan and execute a multi-step task with confirmation gates" — landing in Phase 3 as Chuck Norris autonomous mode.
aish is not a wrapper around bash. It's a first-class interactive environment where the shell is one of several execution channels.
Status
| Component | State |
|---|---|
| Repository skeleton | ✅ in this commit |
| Phase 0 manifest | ✅ docs/PHASE0.md — locked |
| Phase 0 implementation | 🔜 next session |
| Phase 1+ | 📋 enumerated in PHASE0.md §11 |
Every module file currently raises not implemented (Phase 0 pending)
when called. luajit main.lua fails loudly at the first un-implemented
function, never silently.
Quick orientation
| Read this | If you want to know |
|---|---|
docs/PHASE0.md §1–2 |
What aish is and what Phase 0 ships |
docs/PHASE0.md §3 |
Technology decisions (LuaJIT, FFI, readline, libcurl, llama.cpp) |
docs/PHASE0.md §4 |
Directory layout — these file names are stable across all phases |
docs/PHASE0.md §5 |
How input is dispatched (meta / shell / AI) |
docs/PHASE0.md §6 |
Broker contract: /v1/chat/completions, CMD: extraction |
docs/PHASE0.md §10 |
Config schema and resolution order |
docs/PHASE0.md §11 |
Phase sequence (what lands when) |
docs/PHASE0.md §13 |
Open questions, tracked per phase |
CLAUDE.md |
Project conventions for AI-assisted contributors |
Directory layout
aish/
├── main.lua # entry point
├── repl.lua # readline loop, dispatch, prompt
├── broker.lua # llama.cpp HTTP client
├── router.lua # input classifier (meta/shell/AI)
├── executor.lua # command exec + CMD: extraction
├── context.lua # in-memory turn history
├── history.lua # disk persistence (Phase 1+)
├── safety.lua # destructive-op gate (Phase 3+)
├── renderer.lua # output formatting
├── config.lua # default model registry + preferences
├── ffi/
│ ├── curl.lua # libcurl easy interface
│ ├── readline.lua # GNU readline
│ ├── pty.lua # forkpty (Phase 1+)
│ └── libc.lua # chdir, errno, strerror
└── docs/
└── PHASE0.md # locked substrate
Build / runtime dependencies
System packages (Debian / ALARM / Arch names):
luajit(>= 2.0)libcurl4/libcurl-openssl-3runtimelibreadline8runtimelibc6runtime (always present)
No compilation, no luarocks, no make. Just luajit main.lua.
Running
Once Phase 0 ships:
luajit main.lua # uses ~/.config/aish/config.lua
luajit main.lua --config ./config.lua # explicit config path
AISH_CONFIG=/path/to/config.lua luajit main.lua
Config resolution order is documented in docs/PHASE0.md §10.
Configuration
config.lua is a Lua file returning a single table. The committed
config.lua in this repo is both the canonical example and the
development-fallback config (lowest precedence). Copy it to
~/.config/aish/config.lua and edit endpoints to your local llama.cpp
servers, or point AISH_CONFIG at your own.
The default endpoints assume mfritsche's home network:
fast→dirac.fritz.box:8081(Qwen2.5-Coder-7B q4 8k ctx)deep→dirac.fritz.box:8080(Qwen2.5-Coder-7B q4 32k ctx)cloud→hossenfelder.fritz.box:8082(forwards to OpenRouter)
Replace these with your own llama.cpp endpoints if you're not on that LAN.
License
Not yet selected. Default-private until decided.
Project conventions
See CLAUDE.md for contribution conventions, commit style,
and the phase-loop discipline this project follows.