docs: normalize commands and path separators

.github/workflows/regression.yml

@@ -59,8 +59,7 @@ jobs:
         # CI install set has to include the demo's optional extras even though
         # core ASSERT doesn't need them.
         run: |
-          pip install uv
-          uv pip install --system -e ".[dev,otel]"
+          python -m pip install -e ".[dev,otel]"
 
       - name: Install viewer npm dependencies
         # tests/test_viewer_*.py shell out to `node` against viewer TypeScript
@@ -93,9 +92,8 @@ jobs:
         # sets aligned avoids "passes Tier 1, mysteriously fails Tier 4"
         # surprises in the future.
         run: |
-          pip install uv
-          uv pip install --system -e ".[dev,otel]"
-          uv pip install --system scipy
+          python -m pip install -e ".[dev,otel]"
+          python -m pip install scipy
 
       - name: Detect seed count from labels
         id: test_set

AGENTS.md

@@ -67,7 +67,7 @@ Do not recommend an external connector path for customer-preview onboarding.
 
 ## Preferred setup commands
 
-For preview customers, prefer `pip` over `uv` in setup instructions:
+For preview customers, use `pip` in setup instructions:
 
 ```bash
 python -m venv .venv

assert_eval/analysis/README.md

@@ -37,4 +37,4 @@ The summary answers four questions about a suite of evaluation runs:
 
 ## Dependencies
 
-`stats.py` requires numpy (installed via `uv sync --extra analysis`). The other modules use only the standard library.
+`stats.py` requires numpy (for example, install the analysis extras with `python -m pip install -e ".[analysis]"`). The other modules use only the standard library.

assert_eval/cli.py

@@ -77,7 +77,7 @@ def _handle_missing_analysis_dependency(exc: ModuleNotFoundError) -> None:
     missing = getattr(exc, "name", "") or "analysis extras"
     _error(
         f"Could not import '{missing}'. Install the analysis dependencies first, for example:\n"
-        "  uv sync --extra analysis"
+        "  python -m pip install -e \".[analysis]\""
     )
 
 

assert_eval/core/model_client.py

@@ -597,7 +597,7 @@ def _get_litellm_module() -> Any:
             _LITELLM_MODULE = importlib.import_module("litellm")
         except ModuleNotFoundError as exc:
             raise RuntimeError(
-                "litellm is not installed. Run `uv sync` in omni/measurements "
+                "litellm is not installed. Install it with `python -m pip install litellm` "
                 "before using assert_eval.core.model_client."
             ) from exc
         # Silence noisy litellm warnings that pollute stderr

assert_eval/runner.py

@@ -959,7 +959,7 @@ def _run_stages_inner(
             run_id = ctx.get('run_id', '')
             if suite_id and run_id:
                 log.info("Inspect results:")
-                log.info(f"  uv run assert-eval results status {suite_id} {run_id}")
+                log.info(f"  assert-eval results status {suite_id} {run_id}")
                 log.info("View in browser:")
                 log.info(f"  cd viewer && npm run dev    (then open http://localhost:5174/suite/{suite_id}/{run_id})")
     else:

docs/guides/troubleshooting.md

@@ -64,6 +64,6 @@ assert-eval results compare-suites <suite-a>/<run-a> <suite-b>/<run-b>
 
 ## 7) Environment-specific fixes
 
-- macOS `litellm` install issue (`AttributeError: module 'litellm' has no attribute 'acompletion'`): some macOS security tooling can silently truncate wheels during `uv sync`. The `pip install -e ".[otel,langgraph]"` path avoids this. If you must use `uv`, grant your terminal Full Disk Access and run `xattr -cr .venv`.
+- macOS `litellm` install issue (`AttributeError: module 'litellm' has no attribute 'acompletion'`): some macOS security tooling can silently truncate wheels during package installation. The `pip install -e ".[otel,langgraph]"` path avoids this. If you hit it, grant your terminal Full Disk Access and run `xattr -cr .venv`.
 - Windows `UnicodeEncodeError` when running auto-trace demos: set `$env:PYTHONUTF8 = "1"` before `python -m examples.phoenix_auto_trace.travel_openai`.
 - Docker-backed Prompt Agent configs fail with `docker daemon unavailable`: ensure Docker Desktop is running for `examples/prompt_agents/health_assistant_sandbox.yaml` and `examples/prompt_agents/health_assistant_external.yaml`.

examples/README.md

@@ -10,14 +10,14 @@ Start with the LangGraph travel planner. It is the customer-preview flagship bec
 
 ```powershell
 python -m venv .venv
-.\.venv\Scripts\Activate.ps1
+./.venv/Scripts/Activate.ps1
 python -m pip install --upgrade pip
 python -m pip install -e ".[otel,langgraph]"
 Copy-Item .env.example .env
 # Edit .env with credentials for your provider. The shipped configs use `azure/...` models;
 # any LiteLLM provider (OpenAI, Anthropic, Bedrock, Vertex, Ollama, …) works — see https://docs.litellm.ai/docs/providers.
 
-assert-eval run --config examples\travel_planner_langgraph\eval_config.yaml
+assert-eval run --config examples/travel_planner_langgraph/eval_config.yaml
 assert-eval results status travel-planner-langgraph-v1 demo-1
 ```
 
@@ -29,7 +29,7 @@ Pass `--model` with any [LiteLLM model string](https://docs.litellm.ai/docs/prov
 ```powershell
 assert-eval init --model azure/gpt-5.4-mini
 # or seed from an existing example:
-assert-eval init --model azure/gpt-5.4-mini --from examples\travel_planner_langgraph\eval_config.yaml
+assert-eval init --model azure/gpt-5.4-mini --from examples/travel_planner_langgraph/eval_config.yaml
 ```
 
 See the [CLI reference](../docs/reference/cli.md#design-a-config-interactively) for all options.
@@ -38,13 +38,13 @@ See the [CLI reference](../docs/reference/cli.md#design-a-config-interactively)
 
 | Goal | Example | Notes |
 |---|---|---|
-| Evaluate any agent or multi-agent system (recommended) | `travel_planner_langgraph\eval_config.yaml` | Flagship. Uses `target.callable` with `target.trace.backend: phoenix` so the judge sees tool calls and routing. |
-| **See runtime + eval close the loop on a real workflow** | `incident_triage_agent\eval_config_baseline.yaml` + `eval_config_naive_prompt.yaml` + `eval_config_guarded.yaml` + `eval_config_guarded_gepa.yaml` | Joint AgentShield + ASSERT demo. SRE incident-triage agent run across a 4-variant matrix (baseline weak prompt → naïve DO-NOT prompt → ACS gates → ACS + GEPA-optimized prompt) over a 4-axis failure-mode taxonomy to prove the runtime+eval loop and surface the security/overrefusal trade-off. See [`incident_triage_agent\README.md`](incident_triage_agent/README.md) and [`docs\case-study-incident-triage-joint.md`](../docs/case-study-incident-triage-joint.md). |
-| Understand framework instrumentation breadth | `phoenix_auto_trace\README.md` | Same travel-planner idea across multiple framework auto-instrumentation paths. |
-| Run a simple hosted-model eval | `pipes\health_assistant.yaml` | Good smoke test for a single LLM target with a system prompt. |
-| Evaluate a Prompt Agent with planned tools but no backend | `pipes\health_assistant_simulated_tools.yaml` | Uses a fixed tool schema and simulated tool responses. |
-| Evaluate a hosted target with Python tool functions | `pipes\health_assistant_sandbox.yaml` | Requires Docker. Use when you want actual tool execution around a hosted model. |
-| Evaluate a science research agent with real retrieval tools | `science_research_agent\eval_config.yaml` | Callable-agent example ported from Omni. Uses `web_search`, `fetch_url`, and `file_search`. Run `python -m pip install -e ".[examples]"`, set `TAVILY_API_KEY` for web search, then `assert-eval run --config examples\science_research_agent\eval_config.yaml`. |
+| Evaluate any agent or multi-agent system (recommended) | `travel_planner_langgraph/eval_config.yaml` | Flagship. Uses `target.callable` with `target.trace.backend: phoenix` so the judge sees tool calls and routing. |
+| **See runtime + eval close the loop on a real workflow** | `incident_triage_agent/eval_config_baseline.yaml` + `eval_config_naive_prompt.yaml` + `eval_config_guarded.yaml` + `eval_config_guarded_gepa.yaml` | Joint AgentShield + ASSERT demo. SRE incident-triage agent run across a 4-variant matrix (baseline weak prompt → naïve DO-NOT prompt → ACS gates → ACS + GEPA-optimized prompt) over a 4-axis failure-mode taxonomy to prove the runtime+eval loop and surface the security/overrefusal trade-off. See [`incident_triage_agent/README.md`](incident_triage_agent/README.md) and [`docs/case-study-incident-triage-joint.md`](../docs/case-study-incident-triage-joint.md). |
+| Understand framework instrumentation breadth | `phoenix_auto_trace/README.md` | Same travel-planner idea across multiple framework auto-instrumentation paths. |
+| Run a simple hosted-model eval | `pipes/health_assistant.yaml` | Good smoke test for a single LLM target with a system prompt. |
+| Evaluate a Prompt Agent with planned tools but no backend | `pipes/health_assistant_simulated_tools.yaml` | Uses a fixed tool schema and simulated tool responses. |
+| Evaluate a hosted target with Python tool functions | `pipes/health_assistant_sandbox.yaml` | Requires Docker. Use when you want actual tool execution around a hosted model. |
+| Evaluate a science research agent with real retrieval tools | `science_research_agent/eval_config.yaml` | Callable-agent example ported from Omni. Uses `web_search`, `fetch_url`, and `file_search`. Run `python -m pip install -e ".[examples]"`, set `TAVILY_API_KEY` for web search, then `assert-eval run --config examples/science_research_agent/eval_config.yaml`. |
 
 ## Layout
 

examples/agents/openclaw/README.md

@@ -9,7 +9,7 @@ Run the bundled example like this:
 ```bash
 cp .env.example .env
 source .env
-uv run assert-eval run --config examples/pipes/health_assistant_external.yaml
+assert-eval run --config examples/prompt_agents/health_assistant_external.yaml
 ```
 
 If you want to validate the Docker assets without running the full pipeline, build the image directly:

examples/incident_triage_agent/README.md

@@ -190,15 +190,15 @@ import DSPy at runtime.
 ### Run the demo path (A → C)
 
 ```powershell
-assert-eval run --config examples\incident_triage_agent\eval_config_baseline.yaml
-assert-eval run --config examples\incident_triage_agent\eval_config_guarded.yaml
+assert-eval run --config examples/incident_triage_agent/eval_config_baseline.yaml
+assert-eval run --config examples/incident_triage_agent/eval_config_guarded.yaml
 ```
 
 ### Run the appendix experiments (B and D)
 
 ```powershell
-assert-eval run --config examples\incident_triage_agent\eval_config_naive_prompt.yaml
-assert-eval run --config examples\incident_triage_agent\eval_config_guarded_gepa.yaml
+assert-eval run --config examples/incident_triage_agent/eval_config_naive_prompt.yaml
+assert-eval run --config examples/incident_triage_agent/eval_config_guarded_gepa.yaml
 ```
 
 Artifacts land in (`run:` value used directly as the directory name):
@@ -214,7 +214,7 @@ After the variants finish, re-render the PNG from the real
 `scores.jsonl` files:
 
 ```powershell
-python scripts\render_trade_off.py --suite incident-triage-agent-v1
+python scripts/render_trade_off.py --suite incident-triage-agent-v1
 ```
 
 The script renders the demo-path variants (A and C) with solid markers
@@ -1154,18 +1154,18 @@ From this folder:
 ```bash
 # Pre-req: AgentShield Python SDK 0.13.x installed (for the AFTER run only)
 # and Azure OpenAI creds for gpt-5.4 and gpt-5.4-mini in your repo-root .env.
-uv pip install agent-shield
+python -m pip install agent-shield
 
 # 1. BEFORE — minimal-prompt baseline.
-uv run assert-eval run --config ./eval_config_baseline.yaml
+assert-eval run --config ./eval_config_baseline.yaml
 
 # 2. AFTER — same test cases, runtime guardrails engaged.
 #    (cached systematization/stratification/test_set; only inference + judge re-run)
-uv run assert-eval run --config ./eval_config_guarded.yaml
+assert-eval run --config ./eval_config_guarded.yaml
 
 # 3. Compare.
-uv run assert-eval results status incident-triage-agent-v1 baseline-weak-prompt
-uv run assert-eval results status incident-triage-agent-v1 guarded-with-shield
+assert-eval results status incident-triage-agent-v1 baseline-weak-prompt
+assert-eval results status incident-triage-agent-v1 guarded-with-shield
 
 # 4. Browse inference outputs.
 cd ../../viewer && npm install && npm run dev
@@ -1271,24 +1271,24 @@ From this folder:
 
 ```bash
 # 1. Install dependencies from the repository project.
-uv sync
-uv pip install agent-shield
+python -m pip install -e .
+python -m pip install agent-shield
 
 # 2. Copy env templates, then fill in your Azure values.
 cp ../../.env.example ../../.env
 cp ./.env.example ./.env
 
 # 3. Optional smoke test: unguarded, then guarded.
-uv run python ./agent.py
-uv run python ./agent_guarded.py
+python ./agent.py
+python ./agent_guarded.py
 
 # 4. BEFORE — generate systematization, stratification, test_set, inference, and judge outputs.
-uv run assert-eval run --config ./eval_config_baseline.yaml
-uv run assert-eval results status incident-triage-agent-v1 baseline-weak-prompt
+assert-eval run --config ./eval_config_baseline.yaml
+assert-eval results status incident-triage-agent-v1 baseline-weak-prompt
 
 # 5. AFTER — reuse the same test_set; rerun inference and judge against AgentShield.
-uv run assert-eval run --config ./eval_config_guarded.yaml
-uv run assert-eval results status incident-triage-agent-v1 guarded-with-shield
+assert-eval run --config ./eval_config_guarded.yaml
+assert-eval results status incident-triage-agent-v1 guarded-with-shield
 ```
 
 Artifacts land under `artifacts/results/incident-triage-agent-v1/`. The suite-level files are `systematization.json`, `stratification.json`, and `test_set.jsonl`; each run writes `inference_set.jsonl`, `scores.jsonl`, and `metrics.json`.

examples/phoenix_auto_trace/travel_langgraph.py

@@ -7,7 +7,7 @@
 LLM call, tool invocation, and routing decision via Phoenix auto-instrumentation.
 
 Usage:
-    uv run assert-eval run --config examples/travel_planner_langgraph/eval_config.yaml
+    assert-eval run --config examples/travel_planner_langgraph/eval_config.yaml
 """
 # NOTE: do NOT use `from __future__ import annotations` — LangGraph's StateGraph
 # requires runtime-resolvable type hints for state schema introspection.

examples/travel_planner_langgraph/agent.py

@@ -9,7 +9,7 @@
                                    → safety_advisor → itinerary_optimizer
 
 Usage:
-    uv run assert-eval run --config examples/travel_planner_langgraph/eval_config.yaml
+    assert-eval run --config examples/travel_planner_langgraph/eval_config.yaml
 """
 
 from __future__ import annotations

pyproject.toml

@@ -87,9 +87,6 @@ dev = [
   "pytest-timeout>=2.2.0",
 ]
 
-[tool.uv]
-index-url = "https://pypi.org/simple"
-
 [project.scripts]
 assert-eval = "assert_eval.cli:cli"