refactor: rename env vars to PYAUTO_* prefix

.claude/skills/run_all_script_fix_failures/SKILL.md

@@ -20,7 +20,7 @@ Match PyAutoBuild behavior instead of inventing a separate policy:
 3. Use test mode by default:
 
 ```bash
-PYAUTOFIT_TEST_MODE=1 python <script.py>
+PYAUTO_TEST_MODE=1 python <script.py>
 ```
 
 4. Build the script list using the PyAutoBuild skip policy and the workspace ordering above. Do not use `run_all_scripts.sh`, because that workflow is designed around continuing after failures and writing failure logs under `failed/`.

.github/copilot-instructions.md

@@ -12,7 +12,7 @@ You are working on the **autogalaxy_workspace**, a tutorial/example repository f
 
 ## Testing
 
-`run_all_scripts.sh` sets `PYAUTOFIT_TEST_MODE=1` automatically. Every script should pass in this mode. A script that fails in test mode indicates a real problem (broken import, wrong function name, etc.).
+`run_all_scripts.sh` sets `PYAUTO_TEST_MODE=1` automatically. Every script should pass in this mode. A script that fails in test mode indicates a real problem (broken import, wrong function name, etc.).
 
 ## Notebook Generation
 

AGENTS.md

@@ -20,15 +20,15 @@ python scripts/imaging/modeling.py
 
 ## Testing
 
-Set `PYAUTOFIT_TEST_MODE=1` so non-linear searches skip actual sampling (fast test run).
+Set `PYAUTO_TEST_MODE=1` so non-linear searches skip actual sampling (fast test run).
 
 Run all scripts:
 
 ```
 bash run_all_scripts.sh
 ```
 
-This sets `PYAUTOFIT_TEST_MODE=1` automatically, runs every script in `scripts/`, and reports pass/fail/skipped counts. Failure logs are written to `failed/<relative_path>.log`.
+This sets `PYAUTO_TEST_MODE=1` automatically, runs every script in `scripts/`, and reports pass/fail/skipped counts. Failure logs are written to `failed/<relative_path>.log`.
 
 ### Script ordering
 

CLAUDE.md

@@ -22,19 +22,19 @@ python scripts/interferometer/modeling.py
 
 Jupyter notebooks mirror the scripts in `notebooks/` and use `# %%` markers to separate cells.
 
-**Integration testing / fast mode**: Set `PYAUTOFIT_TEST_MODE=1` to run any example script without performing full non-linear search sampling. The search is skipped and a mock result is returned immediately, making this ideal for high-level integration testing of scripts end-to-end:
+**Integration testing / fast mode**: Set `PYAUTO_TEST_MODE=1` to run any example script without performing full non-linear search sampling. The search is skipped and a mock result is returned immediately, making this ideal for high-level integration testing of scripts end-to-end:
 
 ```bash
-PYAUTOFIT_TEST_MODE=1 python scripts/imaging/modeling.py
+PYAUTO_TEST_MODE=1 python scripts/imaging/modeling.py
 ```
 
-**Fast smoke tests**: For maximum speed, combine test mode with small datasets and fast plots:
+**Fast smoke tests**: For maximum speed, combine test mode with all skip flags and small datasets:
 
 ```bash
-PYAUTOFIT_TEST_MODE=2 PYAUTO_WORKSPACE_SMALL_DATASETS=1 PYAUTO_FAST_PLOTS=1 python scripts/imaging/modeling.py
+PYAUTO_TEST_MODE=2 PYAUTO_SKIP_FIT_OUTPUT=1 PYAUTO_SKIP_VISUALIZATION=1 PYAUTO_SKIP_CHECKS=1 PYAUTO_SMALL_DATASETS=1 PYAUTO_FAST_PLOTS=1 python scripts/imaging/modeling.py
 ```
 
-- `PYAUTO_WORKSPACE_SMALL_DATASETS=1` — caps all grids/masks to 15x15 pixels at 0.6"/px, making simulators and all downstream computations dramatically faster. Delete `dataset/` when toggling this variable.
+- `PYAUTO_SMALL_DATASETS=1` — caps all grids/masks to 15x15 pixels at 0.6"/px, making simulators and all downstream computations dramatically faster. Delete `dataset/` when toggling this variable.
 - `PYAUTO_FAST_PLOTS=1` — skips `plt.tight_layout()` and critical curve/caustic overlay computation in subplot functions.
 
 **Codex / sandboxed runs**: when running from Codex or any restricted environment, set writable cache directories so `numba` and `matplotlib` do not fail on unwritable home or source-tree paths:
@@ -55,14 +55,14 @@ Run all scripts end-to-end using the provided runner:
 bash run_all_scripts.sh
 ```
 
-- Sets `PYAUTOFIT_TEST_MODE=1` automatically (skips non-linear sampling)
+- Sets `PYAUTO_TEST_MODE=1` automatically (skips non-linear sampling)
 - Failure logs written to `failed/<relative_path>.log`; execution continues after each failure
 - Final summary shows pass/fail/skipped counts
 
 To run a single script in test mode:
 
 ```bash
-PYAUTOFIT_TEST_MODE=1 python3 scripts/imaging/modeling.py
+PYAUTO_TEST_MODE=1 python3 scripts/imaging/modeling.py
 ```
 
 ### Script ordering rules

config/build/env_vars.yaml

@@ -9,8 +9,11 @@
 #   - Patterns without '/' match the file stem exactly
 
 defaults:
-  PYAUTOFIT_TEST_MODE: "2"              # 0=normal, 1=reduced iterations, 2=skip sampler (fastest)
-  PYAUTO_WORKSPACE_SMALL_DATASETS: "1"  # Cap grids/masks to 15x15, reduce MGE gaussians
+  PYAUTO_TEST_MODE: "2"                 # 0=normal, 1=reduced iterations, 2=skip sampler (fastest)
+  PYAUTO_SKIP_FIT_OUTPUT: "1"           # Skip pre/post-fit I/O, VRAM profiling, result text
+  PYAUTO_SKIP_VISUALIZATION: "1"        # Skip fit visualization and plotting
+  PYAUTO_SKIP_CHECKS: "1"              # Skip mesh validation, position checks, weight thresholds
+  PYAUTO_SMALL_DATASETS: "1"           # Cap grids/masks to 15x15, reduce MGE gaussians
   PYAUTO_DISABLE_JAX: "1"              # Force use_jax=False, avoid JIT compilation overhead
   PYAUTO_FAST_PLOTS: "1"               # Skip tight_layout() + critical curve/caustic overlays
   JAX_ENABLE_X64: "True"               # Enable 64-bit precision in JAX
@@ -20,10 +23,10 @@ defaults:
 overrides:
   # Tutorial loads FITS data at full resolution
   - pattern: "howtogalaxy/chapter_1_introduction/tutorial_3_fitting"
-    unset: [PYAUTO_WORKSPACE_SMALL_DATASETS]
+    unset: [PYAUTO_SMALL_DATASETS]
   # guides/results/start_here.py must produce real samples so the example
   # scripts that read from `output/results_folder` afterwards
   # (data_fitting, queries, models, samples_via_aggregator) find a
   # non-empty aggregator. Covers all scripts under guides/results/.
   - pattern: "guides/results/"
-    unset: [PYAUTOFIT_TEST_MODE]
+    unset: [PYAUTO_TEST_MODE, PYAUTO_SKIP_FIT_OUTPUT]

run_all_scripts.sh

@@ -1,5 +1,5 @@
 #!/bin/bash
-# Run all scripts in scripts/ with PYAUTOFIT_TEST_MODE=1.
+# Run all scripts in scripts/ with PYAUTO_TEST_MODE=1.
 #
 # Rules:
 #   - start_here.py in a folder runs before all other scripts and subfolders in that folder
@@ -12,7 +12,10 @@ FAILED_DIR="$SCRIPT_DIR/failed"
 NO_RUN_YAML="$SCRIPT_DIR/../PyAutoBuild/autobuild/config/no_run.yaml"
 PROJECT_KEY="autogalaxy"
 
-export PYAUTOFIT_TEST_MODE=1
+export PYAUTO_TEST_MODE=1
+export PYAUTO_SKIP_FIT_OUTPUT=1
+export PYAUTO_SKIP_VISUALIZATION=1
+export PYAUTO_SKIP_CHECKS=1
 
 # ---------------------------------------------------------------------------
 # Build ordered script list: within each directory, start_here.py runs first,

scripts/guides/results/start_here.py

@@ -93,9 +93,9 @@
 
 model = af.Collection(galaxies=af.Collection(galaxy=galaxy))
 
-test_mode_was_on = os.environ.get("PYAUTOFIT_TEST_MODE") == "1"
+test_mode_was_on = os.environ.get("PYAUTO_TEST_MODE") == "1"
 if test_mode_was_on:
-    os.environ.pop("PYAUTOFIT_TEST_MODE", None)
+    os.environ.pop("PYAUTO_TEST_MODE", None)
 
 search = af.Nautilus(
     path_prefix=Path("results_folder"),
@@ -111,7 +111,7 @@
 result = search.fit(model=model, analysis=analysis)
 
 if test_mode_was_on:
-    os.environ["PYAUTOFIT_TEST_MODE"] = "1"
+    os.environ["PYAUTO_TEST_MODE"] = "1"
 
 """
 __Info__

smoke_tests.txt

@@ -1,6 +1,6 @@
 imaging/start_here.py
 imaging/fit.py
-# imaging/modeling.py  # disabled: result.max_log_likelihood_instance fails in PYAUTOFIT_TEST_MODE=2 (rhayes777/PyAutoFit#1179)
+# imaging/modeling.py  # disabled: result.max_log_likelihood_instance fails in PYAUTO_TEST_MODE=2 (rhayes777/PyAutoFit#1179)
 interferometer/start_here.py
 multi/start_here.py
 # ellipse/modeling.py  # disabled: bypass mode tuple-path KeyError on ellipse models (rhayes777/PyAutoFit#1179)