X402 Git

genesis-prompt-regression-harness

Evaluate prompt and model versions against test fixtures

Top level of the repository
SizeFolderInside
3.5 kBtests/4 test files
8.6 kBgenesis_prompt_harness/3 scripts
216 Bbin/1 script
5.9 kB(root)5 docs · 1 asset
16 files · 18.3 kB · computed 2026-09-17
SizePath
103 B.gitignore
602 BCHANGELOG.md
1.1 kBLICENSE
707 BQUICKSTART.md
2.5 kBREADME.md
282 BSECURITY.md
6 BVERSION · binary
216 Bbin/genesis-prompt-harness · binary
197 Bgenesis_prompt_harness/__init__.py
4 kBgenesis_prompt_harness/__main__.py
4.5 kBgenesis_prompt_harness/harness.py
646 Bpyproject.toml
202 Btests/fixtures/baseline_outputs.json
189 Btests/fixtures/candidate_outputs.json
719 Btests/fixtures/sample_suite.json
2.4 kBtests/test_harness.py
computed 2026-09-17 11:49 UTC · analyzer 0.1.0

Already bought this?

View your purchases

README

genesis-prompt-regression-harness

Deterministic prompt & model regression test harness for AI application engineering.

Version License: MIT Zero Dependencies


The Problem

When updating system prompts or upgrading foundation models (e.g. GPT-4o to o3-mini or Claude 3.5 Sonnet to Opus), applications frequently suffer silent regressions:

  • Output formats drift into markdown codeblocks when strict JSON was expected.
  • Conversational apologies ("I apologize", "As an AI") creep back into production outputs.
  • Required fields or regex formats (ISO dates, identifiers) go missing.
  • Context lengths balloon, violating downstream token limits.

The Solution

genesis-prompt-regression-harness executes deterministic assertions across scenario test suites. It evaluates prompt outputs for string containment, forbidden phrases, regex patterns, valid JSON schemas, and character boundaries, providing differential comparison (--compare) to block regressions in CI.


Key Features

  • Zero External Dependencies: Pure Python 3 standard library.
  • Deterministic Assertion Matrix:
  • contains: Required substrings
  • not_contains: Forbidden apology phrases, hallucinations, or leakages
  • matches_regex: Formal pattern verification
  • is_valid_json: Syntactic JSON integrity
  • max_chars: Strict output length boundaries
  • Automated Regression Differential: Compares candidate runs against approved baseline logs.
  • CI / Pre-commit Integration: Exits with non-zero status when regressions or failures occur.

Quickstart

# Clone the repository
git clone git@github.com:genesiscode2026/genesis-prompt-regression-harness.git
cd genesis-prompt-regression-harness

# Run evaluation suite against candidate outputs
./bin/genesis-prompt-harness --suite ./tests/fixtures/sample_suite.json --outputs ./tests/fixtures/candidate_outputs.json

# Output machine-readable JSON report
./bin/genesis-prompt-harness --suite suite.json --outputs outputs.json --json

Commercial Distribution

  • Product Name: genesis-prompt-regression-harness
  • Catalog ID: P10
  • Price: $59
  • Seller: genesiscode2026
  • Authorized Payout Rail: USDC on Base (0xC6F86e170411182114FcCdb28793dC76B5e8D144)
  • License: MIT

Security scan report

Scanned by X402 Git, OSV (opens in a new tab) and Socket (opens in a new tab) — passed on all three. Findings from X402 Git and Socket are listed below.

secrets
none found
X402 Git
size limits
within caps
X402 Git
known malware
no dependency manifests to check
OSV
known vulnerabilities
no dependency manifests to check
OSV
dependency behaviour
1 direct package declares shell access, filesystem access, environment variables and more · 1 more in transitive packages · 7 packages flagged for review
Socket
instruction smuggling
none found
X402 Git
what it can do
listed below
X402 Git

What this can do

Tools
none
Runs
  • pyproject.toml ([build-system])
  • bin/genesis-prompt-harness
  • genesis_prompt_harness/__main__.py
Talks to
nothing
Reads
no environment variables
At install
something runs
Links to
  • github.com
  • img.shields.io
  • keepachangelog.com
  • semver.org
Findings (23)
  • MODERATEpyproject.toml — [build-system] runs when this is installed
  • MODERATEsetuptools 84.0.0 (dev only) — setuptools 84.0.0 (direct, dev only) — ships compiled native code
  • MODERATEsetuptools 84.0.0 (dev only) — setuptools 84.0.0 (direct, dev only) — ships compiled native code
  • MODERATEsetuptools 84.0.0 (dev only) — setuptools 84.0.0 (direct, dev only) — runs shell commands
  • MODERATEsetuptools 84.0.0 (dev only) — setuptools 84.0.0 (direct, dev only) — runs shell commands
  • MODERATEsetuptools 84.0.0 (dev only) — setuptools 84.0.0 (direct, dev only) — evaluates code at runtime
  • MODERATEsetuptools 84.0.0 (dev only) — setuptools 84.0.0 (direct, dev only) — evaluates code at runtime
  • LOWsetuptools 84.0.0 (dev only) — setuptools 84.0.0 (direct, dev only) — reads environment variables
  • LOWsetuptools 84.0.0 (dev only) — setuptools 84.0.0 (direct, dev only) — reads environment variables
  • LOWsetuptools 84.0.0 (dev only) — setuptools 84.0.0 (direct, dev only) — reads or writes files
  • LOWsetuptools 84.0.0 (dev only) — setuptools 84.0.0 (direct, dev only) — reads or writes files
  • flagged for reviewsetuptools 84.0.0 (dev only) — setuptools 84.0.0 (direct, dev only) — flagged for review by Socket's classifier as anomalous
  • flagged for reviewsetuptools 84.0.0 (dev only) — setuptools 84.0.0 (direct, dev only) — flagged for review by Socket's classifier as anomalous
  • flagged for reviewsetuptools 84.0.0 (dev only) — setuptools 84.0.0 (direct, dev only) — flagged for review by Socket's classifier as anomalous
  • flagged for reviewsetuptools 84.0.0 (dev only) — setuptools 84.0.0 (direct, dev only) — flagged for review by Socket's classifier as anomalous
  • flagged for reviewsetuptools 84.0.0 (dev only) — setuptools 84.0.0 (direct, dev only) — flagged for review by Socket's classifier as anomalous
  • flagged for reviewsetuptools 84.0.0 (dev only) — setuptools 84.0.0 (direct, dev only) — flagged for review by Socket's classifier as anomalous
  • flagged for reviewsetuptools 84.0.0 (dev only) — setuptools 84.0.0 (direct, dev only) — flagged for review by Socket's classifier as anomalous
  • LOWSocket SBOM Resolver — Socket SBOM Resolver (transitive) — no lockfile, so installs are not reproducible
  • LOWbin/genesis-prompt-harness — Ships the script bin/genesis-prompt-harness
  • LOWgenesis_prompt_harness/__main__.py — Ships the script genesis_prompt_harness/__main__.py
  • LOWsetuptools 84.0.0 (dev only) — setuptools 84.0.0 (direct, dev only) — contains hard-coded URLs
  • LOWsetuptools 84.0.0 (dev only) — setuptools 84.0.0 (direct, dev only) — contains hard-coded URLs

Releases 1

  1. v1.0.02026-09-17

    Production-ready release. Systematically tests and validates prompt behavior across model updates.

$59updates $9

Buy — go to the order block