exp-lens-variance-stability
Skill1 starsChanged 8 months agoMITView on GitHub
Create a variance analysis profile assessing whether signals exceed noise and whether results are stable across random seeds. Stability lens answering "Is the signal larger than the noise?
In this repository59 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- arch-lens-c4-containerSkill
- arch-lens-concurrencySkill
- arch-lens-data-lineageSkill
- arch-lens-deploymentSkill
- arch-lens-developmentSkill
- arch-lens-error-resilienceSkill
- arch-lens-module-dependencySkill
- arch-lens-operationalSkill
- arch-lens-process-flowSkill
- arch-lens-repository-accessSkill
- arch-lens-scenariosSkill
- arch-lens-securitySkill
- arch-lens-state-lifecycleSkill
- audit-archSkill
- audit-bugsSkill
- audit-cohesionSkill
- audit-defense-standardsSkill
- audit-frictionSkill
- audit-implSkill
- audit-testsSkill
- design-guardsSkill
- dry-walkthroughSkill
- elaborate-phaseSkill
- exp-lens-benchmark-representativenessSkill
- exp-lens-causal-assumptionsSkill
- exp-lens-comparator-constructionSkill
- exp-lens-error-budgetSkill
- exp-lens-estimand-claritySkill
- exp-lens-exploratory-confirmatorySkill
- exp-lens-fair-comparisonSkill
- exp-lens-governance-riskSkill
- exp-lens-iterative-learningSkill
- exp-lens-measurement-validitySkill
- exp-lens-pipeline-integritySkill
- exp-lens-randomization-blockingSkill
- exp-lens-reproducibility-artifactsSkill
- exp-lens-sensitivity-robustnessSkill
- exp-lens-severity-testingSkill
- exp-lens-unit-interferenceSkill
- exp-lens-validity-threatsSkill
- id-slopSkill
- implement-worktreeSkill
- investigateSkill
- make-arch-diagSkill
- make-experiment-diagSkill
- make-groupsSkill
- make-planSkill
- make-reqSkill
- make-scenariosSkill
- mermaidSkill
- plan-arcSkill
- rectifySkill
- review-approachSkill
- update-architectureSkill
- update-reqsSkill
- update-specsSkill
- verify-diagSkill
- verify-proposed-diagSkill
The file
We split this skill into the rules it gives an agent
11 rules. About 0 are common, 9 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈9
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 17
Focus: Run-to-Run Variance, Seed Sensitivity, Nondeterminism Sources, Confidence Intervals, Noise Floor
- Distinctive: line 30
Do not litter the codebase with useless comments, TODO markers, or explanatory annotations — the skill output and diagram speak for …
- Very specific: line 37
BEFORE creating any diagram, LOAD the /mermaid skill using the Skill tool - this is MANDATORY
- Distinctive: line 49
Look for: seed, random_state, torch.manual_seed, np.random, set_seed, PYTHONHASHSEED
- Distinctive: line 53
Look for: cudnn, benchmark, deterministic, parallel, async, thread, race, order, nondeterministic
- Distinctive: line 57
Look for: n_runs, trials, repeat, replicate, mean, std, confidence, interval, aggregate
- Distinctive: line 61
Look for: std, stderr, confidence, interval, range, median, quartile, bootstrap, error_bar
- Distinctive: line 65
Look for: significant, difference, improvement, margin, effect_size, gap, overlap
- Distinctive: line 88
Use the mermaid skill conventions to create a stochasticity diagram with:
- Very specific: line 107
Write the diagram to: temp/exp-lens-variance-stability/exp_diag_variance_stability_{YYYY-MM-DD_HHMMSS}.md
- Distinctive: line 207
Before creating the diagram, verify:
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Variance Stability Experimental Design Lens
Philosophical Mode: Stability Primary Question: "Is the signal larger than the noise?" Focus: Run-to-Run Variance, Seed Sensitivity, Nondeterminism Sources, Confidence Intervals, Noise Floor
When to Use
- ML experiments with high variance across seeds
- Systems benchmarks with environmental noise
- Results where small differences are claimed as improvements
- User invokes
/exp-lens-variance-stabilityor/make-experiment-diag variance
Critical Constraints
NEVER:
- Modify any source code files
- Do not litter the codebase with useless comments, TODO markers, or explanatory annotations — the skill output and diagram speak for themselves
ALWAYS:
- Count the actual number of independent runs — single-run results must be flagged prominently
- Assess whether claimed improvements exceed the observed standard deviation
- Identify all sources of nondeterminism, not just random seeds
- Report when confidence intervals are absent — absence is a finding, not an omission
- BEFORE creating any diagram, LOAD the
/mermaidskill using the Skill tool - this is MANDATORY
Analysis Workflow
Step 1: Launch Parallel Exploration Subagents
Spawn Explore subagents to investigate:
Random Seed Management
- Find how seeds are set and varied
- Look for: seed, random_state, torch.manual_seed, np.random, set_seed, PYTHONHASHSEED
Nondeterminism Sources
- Find sources of nondeterminism beyond seeds
- Look for: cudnn, benchmark, deterministic, parallel, async, thread, race, order, nondeterministic
Multiple Run Protocol
- Find how many runs are performed per condition
- Look for: n_runs, trials, repeat, replicate, mean, std, confidence, interval, aggregate
Variance Reporting
- Find how variance is reported (if at all)
- Look for: std, stderr, confidence, interval, range, median, quartile, bootstrap, error_bar
Signal-to-Noise Assessment
- Find whether claimed improvements exceed observed variance
- Look for: significant, difference, improvement, margin, effect_size, gap, overlap
Step 2: Build Variance Profile
For each reported result:
- How many independent runs?
- What is the standard deviation across runs?
- Does the claimed improvement exceed the noise floor?
- Are confidence intervals reported?
- What sources of nondeterminism exist beyond seeds?
Build the variance profile.
Step 3: Analyze Signal vs Noise
CRITICAL — Analyze Signal vs Noise: For every claimed improvement:
- Is the improvement magnitude larger than the run-to-run standard deviation?
- Could the ranking of methods change under reruns?
- Are the "best" results cherry-picked from multiple seeds?
Step 4: Create the Diagram
Use the mermaid skill conventions to create a stochasticity diagram with:
Direction: TB (nondeterminism sources flow down through aggregation to reported results)
Subgraphs:
- NONDETERMINISM SOURCES
- VARIANCE AGGREGATION
- REPORTED RESULTS
Node Styling:
stateNodeclass: Nondeterminism sourceshandlerclass: Aggregation methodsoutputclass: Reported resultsgapclass: Unreported variance or single-run resultsdetectorclass: Confidence intervals and statistical testscliclass: Seed management
Step 5: Write Output
Write the diagram to: temp/exp-lens-variance-stability/exp_diag_variance_stability_{YYYY-MM-DD_HHMMSS}.md
Output Template
# Variance Stability Analysis: {Experiment Name}
**Lens:** Variance Stability (Stability)
**Question:** Is the signal larger than the noise?
**Date:** {YYYY-MM-DD}
**Scope:** {What was analyzed}
## Variance Profile
| Experiment | N Runs | Mean | Std | CI | Signal > Noise? |
|------------|--------|------|-----|----|-----------------|
| {experiment} | {n} | {mean} | {std} | {CI or "Not reported"} | {Yes/No/Unclear} |
## Nondeterminism Inventory
| Source | Type | Controlled? | Impact |
|--------|------|-------------|--------|
| {source} | {seed/hardware/async/etc} | {Yes/No/Partial} | {Low/Medium/High} |
## Stochasticity Diagram
```mermaid
%%{init: {'flowchart': {'nodeSpacing': 50, 'rankSpacing': 60, 'curve': 'basis'}}}%%
graph TB
%% CLASS DEFINITIONS %%
classDef cli fill:#1a237e,stroke:#7986cb,stroke-width:2px,color:#fff;
classDef stateNode fill:#004d40,stroke:#4db6ac,stroke-width:2px,color:#fff;
classDef handler fill:#e65100,stroke:#ffb74d,stroke-width:2px,color:#fff;
classDef phase fill:#6a1b9a,stroke:#ba68c8,stroke-width:2px,color:#fff;
classDef newComponent fill:#2e7d32,stroke:#81c784,stroke-width:2px,color:#fff;
classDef output fill:#00695c,stroke:#4db6ac,stroke-width:2px,color:#fff;
classDef detector fill:#b71c1c,stroke:#ef5350,stroke-width:2px,color:#fff;
classDef gap fill:#ff6f00,stroke:#ffa726,stroke-width:2px,color:#000;
classDef integration fill:#c62828,stroke:#ef9a9a,stroke-width:2px,color:#fff;
subgraph NDSources ["NONDETERMINISM SOURCES"]
direction TB
SEED["Random Seed<br/>━━━━━━━━━━<br/>Controlled via<br/>seed management"]
HW["Hardware Variance<br/>━━━━━━━━━━<br/>GPU/CPU ordering<br/>differences"]
ASYNC["Async Operations<br/>━━━━━━━━━━<br/>Thread/process<br/>race conditions"]
end
subgraph Aggregation ["VARIANCE AGGREGATION"]
direction TB
MULTI["Multiple Runs<br/>━━━━━━━━━━<br/>N independent<br/>repetitions"]
CI["Confidence Interval<br/>━━━━━━━━━━<br/>Statistical bounds<br/>on estimates"]
end
subgraph Results ["REPORTED RESULTS"]
direction TB
RESULT["Reported Result<br/>━━━━━━━━━━<br/>Mean ± std<br/>with CI"]
SINGLE["Single-Run Result<br/>━━━━━━━━━━<br/>No variance<br/>reported"]
end
SEED --> MULTI
HW --> MULTI
ASYNC --> MULTI
MULTI --> CI
CI --> RESULT
MULTI --> SINGLE
%% CLASS ASSIGNMENTS %%
class SEED cli;
class HW,ASYNC stateNode;
class MULTI handler;
class CI detector;
class RESULT output;
class SINGLE gap;Seed Sensitivity Assessment
| Seed | Run Result | Rank Among Methods |
| {seed} | {result} | {rank} |
Reporting Completeness Checklist
- Number of runs reported per condition
- Standard deviation or standard error reported
- Confidence intervals reported
- All seeds or seed range disclosed
- Nondeterminism sources acknowledged
Key Findings
- {Description of whether signals exceed noise and reporting completeness}
---
## Pre-Diagram Checklist
Before creating the diagram, verify:
- [ ] LOADED `/mermaid` skill using the Skill tool
- [ ] Using ONLY classDef styles from the mermaid skill (no invented colors)
- [ ] Diagram will include a color legend table
---
## Related Skills
- `/make-experiment-diag` - Parent skill for lens selection
- `/mermaid` - MUST BE LOADED before creating diagram
- `/exp-lens-reproducibility-artifacts` - For environment and artifact reproducibility
- `/exp-lens-error-budget` - For systematic error and bias analysisDoes it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

