nooa-cli 0.0.7__tar.gz

nooa_cli-0.0.7/.gitignore

@@ -0,0 +1,407 @@
+# Rust toolchain (installed for nemo_flow PyO3 build)
+.cargo/
+.rustup/
+
+# Node.js dependencies (never commit)
+node_modules/
+
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+!packages/nemo-oo-agents-viewer/frontend-react/dist/
+!src/nooa/viewer/frontend-react/dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+# Exception: allow util/*/lib/ directories (our code, not build artifacts)
+!util/*/lib/
+!util/**/lib/
+parts/
+sdist/
+var/
+wheels/
+pip-wheel-metadata/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+#  Usually these files are written by a python script from a template
+#  before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.nox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+*.py,cover
+.hypothesis/
+.pytest_cache/
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+db.sqlite3-journal
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# IPython
+profile_default/
+ipython_config.py
+
+# pyenv
+.python-version
+
+# pipenv
+#   According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
+#   However, in case of collaboration, if having platform-specific dependencies or dependencies
+#   having no cross-platform support, pipenv may install dependencies that don't work, or not
+#   install all needed dependencies.
+#Pipfile.lock
+
+# PEP 582; used by e.g. github.com/David-OConnor/pyflow
+__pypackages__/
+
+# Celery stuff
+celerybeat-schedule
+celerybeat.pid
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.env.*
+!.env.example
+!.env.template
+.venv
+.venv-macos
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+
+# uv cache directory
+.uv-cache/
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre/
+
+# IDEs
+.vscode/*
+.vscode-server/*
+.idea/*
+*.swp
+*.swo
+
+# OS
+.DS_Store
+Thumbs.db
+
+# Temporal
+temporal/
+.temporal/
+
+# Custom
+secrets.env
+local_config.yaml
+
+# Security: JWT tokens and API keys
+**/test-nvauth.py
+**/debug-auth*.py
+# JWT token patterns (eyJ...)
+eyJ*
+
+# Third-party dependencies
+3P/
+3p/
+
+# Trash folder for temporary files
+trash/
+
+# Tools directory (generated traces) - only root level
+/tools/
+
+# Agent trace files and results (root level)
+/trace/
+/traces/
+/results/
+trace*.json
+!trace_viewer_config.json
+/tmp/
+
+# NVIDIA OO Agents generated trace and evaluation files (anywhere)
+*.noo-eval.jsonl
+*.006summary.json
+# Exception: Allow trace fixtures for testing/eval
+!evaluation/fixtures/**/*.noo-eval.jsonl
+
+# Interface output markdown (experiment results)
+docs/interface_output/
+
+# Trace viewer config backups
+**/trace_viewer_config.json.backup
+
+# Local config overrides (user-specific, not tracked)
+**/*.local.json
+
+# Bridge watchpoint file (runtime job monitoring, not tracked)
+.bridge_watch.json
+
+# Slurm job output files (runtime artifacts, not tracked)
+*.err
+*.out
+
+# Cursor IDE rules (keep local but don't track)
+.cursorrules
+
+# PDFs
+*.pdf
+packages/.venv/
+
+# Experimental implementations (on experimental-implementations branch)
+composer-packages/
+gpt-packages/
+trace_tui/
+NEW_AGENT_TUI.md
+RUN_NEW_AGENT_TUI.md
+SETUP_QWEN_TUI.md
+TUI_INTEGRATION_SUMMARY.md
+run-new-tui.sh
+test-tui-debug.py
+test-tui.sh
+scratch/ALIGNMENT_SUMMARY.md
+scratch/DESIGN_ALIGNMENT_REPORT.md
+packages/nemo_oo_agents-core/src/nemo_oo_agents/cli.py
+packages/nemo_oo_agents-tui/src/nemo_oo_agents_tui/cli.py
+
+# Prompt optimization experiment results
+util/prompt-optimization/results/*.json
+util/prompt-optimization/results/*.jsonl
+util/prompt-optimization/results/*/*.json
+util/prompt-optimization/results/*/*.jsonl
+util/prompt-optimization/traces/
+util/prompt-optimization/traces_test/
+
+# Eval pipeline experiments and data
+util/e2e_optimization/experiments/
+util/e2e_optimization/src/e2e_optimization/examples/**/*.jsonl
+util/e2e_optimization/src/e2e_optimization/examples/**/results/
+# Exception: allow data_*.jsonl (test datasets)
+!util/e2e_optimization/src/e2e_optimization/examples/**/data_*.jsonl
+util/eval_pipeline/experiments/
+
+# Experiment test result directories
+experiments/tau_bench_test*/
+experiments/**/results/
+
+traces.db
+traces.db-wal
+traces.db-shm
+# Large result files
+experiments/**/results/
+
+# Test files
+test_arxiv_fetch.py
+trigger_librarian_fetch.py
+experiments/evaluation-ablations/*.db
+experiments/evaluation-ablations/test_case_*/
+experiments/evaluation-ablations/test.csv
+experiments/evaluation-ablations/test.json
+agents/**/traces/
+
+# Claude Code settings (user-specific)
+.claude/settings.json
+
+# Claude Code runtime directories (cache, sessions, IDE state, plugins, skills)
+.claude/backups/
+.claude/history.jsonl
+.claude/cache/
+.claude/file-history/
+.claude/ide/
+.claude/plugins/
+.claude/policy-limits.json
+.claude/projects/
+.claude/sessions/
+.claude/shell-snapshots/
+.claude.json
+
+# Agent runtime directory
+.agents/
+
+# Shell history
+.bash_history
+
+# User-level XDG config/local dirs (injected by sandbox/environment)
+.config/
+.local/
+
+# Rust toolchain (installed by rustup for nemo-flow build)
+.cargo/
+.rustup/
+
+# VS Code Server (remote development runtime)
+.vscode-server/
+
+debug_http_logs/
+
+# SWE-bench evaluation logs (generated during test runs)
+**/logs/run_evaluation/
+
+# Trace analyzer batch results
+**/analysis_batch/
+
+# Claude Code local settings (personal, not shared)
+.claude/settings.local.json
+.claude/agents
+
+# Claude Code sandbox artifacts (created at runtime, not project files)
+.srt-settings.json
+# Debug handler dumps (written to cwd on SIGUSR2)
+debug_dump_*.txt
+# NeMo OO project directory (libraries, traces, config)
+.nemo_oo/
+.nooa/
+.nemo_oo_agents/
+# Shell config stubs (empty read-only files injected by sandbox)
+.zshrc
+.zprofile
+.bash_profile
+.bashrc
+.profile
+.gitconfig
+.ripgreprc
+.idea
+.vscode
+# Git internal stubs (root-level sandbox artifacts only)
+/HEAD
+/config
+/hooks
+/objects
+/refs
+# pytest temp directories
+pytest-of-*/
+experiments/evaluation-ablations/task_temp/
+
+# DABStep experiment outputs (top-level scratch directory)
+/dabstep/
+
+# Harbor generated task directories (regenerated via `harbor adapter run`)
+util/harbor/tasks/
+
+# OpenSkills workspace files
+.openskills/
+
+# Local lockfile
+prpm.lock
+
+# Git worktrees
+.worktrees/
+worktrees/
+
+# Bridge session ID file — written by slack bridge into each agent worktree root
+.claude_session_id
+
+sft_data/**/*.jsonl
+
+# Git credentials (must not be committed)
+.git-credentials
+-e
+# Bridge watchpoint file — per-worktree, never committed
+.bridge_watch.json
+
+.claude/backups
+.claude/file-history
+.claude/projects
+.claude/shell-snapshots
+.claude/telemetry
+./claude/sessions
+./claude/cache
+./claude/plugins
+
+.local/
+.config/
+util/harbor/tasks/terminal_bench/
+
+# WTF skills (generated by wtf install-skills)
+.claude/skills/wtf*
+tests/integration/fixtures/*.db
+tests/integration/fixtures/*.lock
+
+# Worktree directories
+.wt-*
+
+# Throwaway benchmark smoke-test configs
+util/harbor/tb1_opus_smoke.yaml
+util/harbor/tb1_ultra_smoke.yaml
+.claude/scheduled_tasks.lock
+.commit_msg
+
+# LoCoMo benchmark data (downloaded on demand, not vendored)
+examples/memory_bench/data/
+
+# Local scratch/working files (never commit)
+.scratch/
+
+# Claude Code local session dir (never commit)
+.claude/

nooa_cli-0.0.7/PKG-INFO

@@ -0,0 +1,52 @@
+Metadata-Version: 2.4
+Name: nooa-cli
+Version: 0.0.7
+Summary: CLI for nemo-oo-agents (`nooa` command: eval runner, trace viewer, config)
+Project-URL: Homepage, https://github.com/NVIDIA-NeMo/labs-OO-Agents
+Project-URL: Repository, https://github.com/NVIDIA-NeMo/labs-OO-Agents
+Project-URL: Issues, https://github.com/NVIDIA-NeMo/labs-OO-Agents/issues
+License: Apache-2.0
+Requires-Python: <3.14,>=3.12
+Requires-Dist: click>=8.1.0
+Requires-Dist: nooa
+Requires-Dist: pyyaml>=6.0
+Provides-Extra: ast
+Requires-Dist: tree-sitter-go>=0.23.0; extra == 'ast'
+Requires-Dist: tree-sitter-javascript>=0.23.0; extra == 'ast'
+Requires-Dist: tree-sitter-python>=0.23.0; extra == 'ast'
+Requires-Dist: tree-sitter-rust>=0.23.0; extra == 'ast'
+Requires-Dist: tree-sitter-typescript>=0.23.0; extra == 'ast'
+Requires-Dist: tree-sitter>=0.23.0; extra == 'ast'
+Provides-Extra: datascience
+Requires-Dist: numpy>=1.24.0; extra == 'datascience'
+Requires-Dist: pandas>=2.0.0; extra == 'datascience'
+Requires-Dist: plotly>=5.0.0; extra == 'datascience'
+Requires-Dist: scikit-learn>=1.3.0; extra == 'datascience'
+Requires-Dist: scipy>=1.10.0; extra == 'datascience'
+Description-Content-Type: text/markdown
+
+# nooa-cli
+
+CLI for [nemo-oo-agents](https://github.com/NVIDIA-NeMo/labs-OO-Agents). Ships the `nooa` command with subcommands for running evaluations, browsing traces, and managing config.
+
+## Install
+
+```bash
+uv add nooa-cli
+
+# ...with numpy/pandas/plotly/scipy/sklearn pre-loaded into the LLM REPL
+uv add "nooa-cli[datascience]"
+```
+
+`nooa-cli` automatically pulls in matching `nemo-oo-agents` (the core framework). The `[datascience]` extra adds libraries the LLM can use in REPL-generated code.
+
+## Usage
+
+```bash
+nooa --help
+nooa start-dev            # launch the trace viewer
+nooa eval ...             # eval pipeline runner
+nooa traces ...           # inspect/manage trace files
+```
+
+See the main repo [README](https://github.com/NVIDIA-NeMo/labs-OO-Agents/blob/main/README.md) for the framework documentation.

nooa_cli-0.0.7/README.md

@@ -0,0 +1,25 @@
+# nooa-cli
+
+CLI for [nemo-oo-agents](https://github.com/NVIDIA-NeMo/labs-OO-Agents). Ships the `nooa` command with subcommands for running evaluations, browsing traces, and managing config.
+
+## Install
+
+```bash
+uv add nooa-cli
+
+# ...with numpy/pandas/plotly/scipy/sklearn pre-loaded into the LLM REPL
+uv add "nooa-cli[datascience]"
+```
+
+`nooa-cli` automatically pulls in matching `nemo-oo-agents` (the core framework). The `[datascience]` extra adds libraries the LLM can use in REPL-generated code.
+
+## Usage
+
+```bash
+nooa --help
+nooa start-dev            # launch the trace viewer
+nooa eval ...             # eval pipeline runner
+nooa traces ...           # inspect/manage trace files
+```
+
+See the main repo [README](https://github.com/NVIDIA-NeMo/labs-OO-Agents/blob/main/README.md) for the framework documentation.

nooa_cli-0.0.7/docs/activity-introspection-design.md

@@ -0,0 +1,75 @@
+# Can the TUI report "where the agent is stopped" via sys.monitoring?
+
+Short answer: **`sys.monitoring` is the wrong tool for this, and you don't need it.**
+A point-in-time "where is the agent right now" probe is answered cheaply and
+safely by `sys._current_frames()` + asyncio task introspection, with **zero**
+changes to the core framework.
+
+## How the TUI and agent actually run (verified)
+
+`tui_application.py::_ensure_agent_loop` spins up a **dedicated asyncio loop on
+its own daemon thread** (`nemo-tui-agent-loop`) via `loop.run_forever()`. The
+agent's `handle()` runs as a task on that loop. prompt_toolkit / the UI runs on
+`MainThread`. So:
+
+- Agent work = one coroutine (`TUIAgent.handle`) on the agent loop, agent thread.
+- The "asker" (a slash command / keybinding) originates on the UI thread.
+
+## Why sys.monitoring doesn't fit
+
+`sys.monitoring` is built for *continuous* event collection (debuggers, coverage,
+profilers), not on-demand snapshots. Concretely, from experiments in this repo:
+
+1. **Callbacks fire on the executing thread, not the asker's thread.** A `LINE`
+   callback runs inline on whatever thread is running the instrumented bytecode.
+   There is no thread-filtering API — you instrument the whole interpreter and
+   disambiguate by reading `threading.current_thread()` *inside* the callback.
+   To answer "where is the agent" you'd have to keep a running
+   `last_frame_by_thread` dict and read the agent's entry from the UI thread.
+   That's a continuously-maintained side channel, not a query.
+
+2. **Cost.** Global `LINE` instrumentation measured **~3.1x** runtime on a tight
+   loop (212ms vs 68ms). `PY_START` and `LINE`+`DISABLE` are ~1.0x — but `DISABLE`
+   permanently de-instruments each location until `restart_events()`, so you lose
+   line resolution after the first hit. There is no cheap always-on LINE feed.
+
+3. **It tells you the *frame*, not the *await*.** The agent is async. When it's
+   "stopped" it's usually suspended at an `await` (LLM call, tool, sleep). The
+   running OS frame is just `BaseEventLoop._run_once`. What you want is the
+   suspended coroutine's stack — which sys.monitoring does not give you.
+
+## What works instead (prototype: `agent_locator.py`)
+
+Two thread-safe, cheap primitives:
+
+- **`sys._current_frames()`** — point-in-time frame for every thread. Safe to
+  call from any thread, O(threads). Good for "what is the agent thread executing
+  right now" when it's running CPU-bound Python.
+
+- **asyncio task introspection** — `asyncio.all_tasks(loop)` + `Task.get_stack()`
+  gives the *suspended await stack* of the agent coroutine, which is what "where
+  is it stopped" usually means. **Caveat:** `asyncio.Task` is not safe to walk
+  from another thread, so the UI thread must hop onto the agent loop first via
+  `run_coroutine_threadsafe(locate_agent(loop), loop)` and read the result back.
+  Frame data is returned as plain dataclasses, so it crosses the thread boundary
+  cleanly. This works even while `handle()` is mid-turn because the snapshot
+  coroutine just queues another callback on the same loop.
+
+Verified live: the probe correctly reports the running task as
+`TUIAgent.handle :: ActorRuntime._call_plan._execute_with_event` and the agent
+thread's current frame.
+
+## Recommendation
+
+- For "where is the agent stopped / what is it doing", ship the
+  `agent_locator` probe. No core changes; the TUI calls
+  `locate_agent_from_other_thread(agent_loop)` from a slash command.
+- The main blind spot: if the agent is blocked *inside a C call or a
+  `run_in_executor` thread* (e.g. an LLM HTTP request), `get_stack()` shows the
+  suspended Python frame at the `await`, which is exactly the right answer
+  ("waiting on the model call"), and `_current_frames()` shows the worker thread
+  if you want the lower-level detail.
+- Only reach for `sys.monitoring` if you later want a *continuous timeline* of
+  what the agent executed (a flight recorder), and accept the perf hit or use
+  `PY_START`/sampled `LINE`+`DISABLE`+`restart_events`. That's a different
+  feature from the on-demand probe you asked about.

nooa_cli-0.0.7/pyproject.toml

@@ -0,0 +1,71 @@
+[project]
+name = "nooa-cli"
+# Version is derived from git tags at build time by uv-dynamic-versioning.
+dynamic = ["version"]
+description = "CLI for nemo-oo-agents (`nooa` command: eval runner, trace viewer, config)"
+license = {text = "Apache-2.0"}
+readme = "README.md"
+requires-python = ">=3.12,<3.14"
+dependencies = [
+    # Floor reflects the actual minimum-compatible core (bump when the cli
+    # starts to require a newer core API). No auto-lockstep — published
+    # wheels declare what they actually need, not what was co-released.
+    "nooa",
+    "click>=8.1.0",
+    # CLI config files
+    "pyyaml>=6.0",
+]
+
+[project.urls]
+Homepage = "https://github.com/NVIDIA-NeMo/labs-OO-Agents"
+Repository = "https://github.com/NVIDIA-NeMo/labs-OO-Agents"
+Issues = "https://github.com/NVIDIA-NeMo/labs-OO-Agents/issues"
+
+[project.scripts]
+nooa = "nooa_cli:main"
+
+[project.entry-points."nooa.skills"]
+"nemo.repo" = "nooa_cli.tools.repo_tools:RepoTools"
+
+[project.optional-dependencies]
+# AST-aware code analysis for RepoTools (tree-sitter grammars).
+# The CLI works without these — RepoTools falls back to regex.
+ast = [
+    "tree-sitter>=0.23.0",
+    "tree-sitter-python>=0.23.0",
+    "tree-sitter-javascript>=0.23.0",
+    "tree-sitter-typescript>=0.23.0",
+    "tree-sitter-go>=0.23.0",
+    "tree-sitter-rust>=0.23.0",
+]
+# Data-science stack pre-loaded into the agent's REPL execution namespace
+# (np, pd, px, go, scipy, sklearn). Soft-imported via try/except; they
+# exist purely so LLM-generated CodeAct cells can use these libs without
+# first failing on a missing import.
+datascience = [
+    "numpy>=1.24.0",
+    "pandas>=2.0.0",
+    "plotly>=5.0.0",
+    "scipy>=1.10.0",
+    "scikit-learn>=1.3.0",
+]
+
+[tool.uv.sources]
+nooa = { workspace = true }
+
+[build-system]
+requires = ["hatchling>=1.26.0", "uv-dynamic-versioning>=0.14.0"]
+build-backend = "hatchling.build"
+
+[tool.hatch.version]
+source = "uv-dynamic-versioning"
+
+[tool.uv-dynamic-versioning]
+vcs = "git"
+style = "pep440"
+bump = true
+metadata = false
+fallback-version = "0.0.6"
+
+[tool.hatch.build.targets.wheel]
+packages = ["src/nooa_cli"]