Borrowing it
Nothing to install: this file belongs to jgiox/goodvibes. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/jgiox/goodvibes/main/.claude/skills/model-regression/SKILL.mdgit clone --depth 1 https://github.com/jgiox/goodvibesWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/jgiox/goodvibes/model-regression)<a href="https://agentmods.dev/skills/jgiox/goodvibes/model-regression"><img src="https://agentmods.dev/badge/skills/jgiox/goodvibes/model-regression/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/jgiox/goodvibes/model-regression"><img src="https://agentmods.dev/badge/skills/jgiox/goodvibes/model-regression.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00106 | $0.00644 |
| Opus 5.5 | $0.00042 | $0.00258 |
| Sonnet 5.5 | $0.00021 | $0.00129 |
| Haiku 4.5 | $0.00011 | $0.00064 |
Grade A, and why
model-regression scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 47 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Model regression gate
A model change is only an improvement if it is measured against a fixed baseline, the same way, before and after. This skill makes that check routine.
1. Baseline first, before touching code
- Find the baseline file (default
metrics/baseline.json). If none exists, create one from the current code before making any change, and commit it on its own. - The baseline must record: each metric name and value, the evaluation data (path plus a content hash or version tag), the random seed(s), the exact command that produced it, the commit SHA, and the date.
- Declare the tolerance for each metric up front (for example "AUC may drop by at most 0.005"). Write it in the baseline file, not in your head.
2. Measure the change the same way
- Run the exact baseline command on the exact same evaluation data and seeds.
- Never change the model and the evaluation data or metric code in the same change. Split them, and re-baseline after an evaluation change.
- When results vary between runs, run at least 3 seeds and report the mean and the worst run, not the best one.
- Never tune on the test set. Tuning uses a validation split; the test set is read once per change.
3. Decide
- Any metric worse than baseline minus tolerance: the change fails. Revert it, or stop and report. Never loosen the tolerance or swap the metric to make it pass.
- All metrics within tolerance and the target metric better: the change passes. Update the baseline file in the same commit, with the new numbers and the reason.
- Improvement smaller than run-to-run noise: report it as "no measurable change", not as an improvement.
4. Keep a regression test
- Add or keep a fast test in the normal test suite that runs the model or scorer on a small frozen fixture and asserts each metric is at least baseline minus tolerance.
- The fixture lives in the repo and never changes silently; changing it is an evaluation change (see step 2).
5. Report
Report in this shape:
| Metric | Baseline | New | Delta | Tolerance | Result |
|---|
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 47 lines · 106 tokens per session scan A 49de83699205
model-regression is a skill published in the GitHub repository jgiox/goodvibes (5 stars, last pushed 6d ago), licensed Apache-2.0. It adds 106 tokens to every session and 644 once invoked, about $0.0004 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-26.
Other skills, from other repositories
agent-platform-model-registry
Agent Platform Model Registry Management. Use when you need to upload, list, describe, update, or delete machine learning models (and their versions) in the Agent Platform Model Registry. Don't use for model training, model deployment to endpoints, or managing non-Agent Platform models.
foundry-config-setup
Resolve missing setup caused by a hardcoded Foundry project endpoint or model in a sample. Use when a sample fails because it uses a placeholder/hardcoded projectendpoint (for example "https://your-project.services.ai.azure.com") or a hardcoded model instead of reading them from the environment.
google-cloud-solution-agentic-analytics-spark-knowledge-catalog
Discovers requirements and designs an end-to-end governed agentic analytics solution using Knowledge Catalog and Managed Service for Apache Spark (Lightning Engine). Use when designing data science and analytics workflows across structured and unstructured distributed data (including in S3, Azure Blob, AlloyDB, and…
training-check
Interactively monitor training metrics from the current Codex session, periodically checking WandB or fallback logs for NaN, divergence, plateaus, and broken runs.
nemo-automodel-launcher-config
Configure NeMo AutoModel job launches for interactive runs, Slurm clusters, and SkyPilot cloud execution.
finetuning
Fine-tune models on Microsoft Foundry using SFT (supervised), DPO (preference), or RFT (reinforcement with graders). Covers dataset preparation, training job submission, deployment, and evaluation. USE FOR: fine-tune, SFT, DPO, RFT, training data, grader, distillation, fine-tuned model, training job, large file…