Augment test
Copier template that bootstraps AI-agent-optimized project setups: layered context routing, curated skills, and an enforced validation gate — synced across projects.
npx -y skills add SamyakJhaveri/loam --skill augment-testAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Data augmentation testing workflow. Use when testing a new augmentation transform, diagnosing transform failures on a specific input, or validating transforms against known-good baselines. Runs augmentation pipeline and reports pass/fail per transform.
SKILL.md
1.7 KB, as published. Nobody here has run it
Augmentation Testing Workflow
Structured workflow for testing data augmentation or transformation pipelines on specific inputs.
Arguments
$ARGUMENTS— input identifier or path to test
Workflow
Phase 1: Run Augmentation
Execute the project's augmentation pipeline on the specified input:
python3 <augmentation-script> <input> \
--level <N> --seed 42 -v
Phase 2: Analyze Results
Check exit status and output for:
- BUILD_FAIL — augmented output doesn't compile/parse
- TRANSFORM_FAIL — transform itself errored
- RUN_FAIL — compiled but crashed at runtime
- VERIFY_FAIL — ran but produced wrong output
- PASS — all good
Phase 3: Diagnose Failures
If the run failed, check against known bugs in .claude/rules/known-issues.md:
- Cross-reference with documented failure patterns
- Check if the input is in a known-failing list
- Identify whether the failure is in the transform or the input
Phase 4: Report
Input: <name>
Augment Level: <N>
Seed: <N>
Result: PASS / FAIL (<type>)
Diagnosis: <known bug reference or new finding>
Transforms Applied: <list or "none">
Phase 5: Escalate (if new bug)
If the failure doesn't match any known bug:
- Document the new bug with reproduction steps
- Add it to
docs/known-failures.md - Create a minimal reproduction case