More room for your next idea. Membership from $9.99/month.Take a look

Debug Evidence Kitv1.0.0

Reproduce a defect, test competing causes, apply a scoped correction and capture fresh before/after evidence. Compare identical commands, changed source and preserved regression tests in a portable interactive case. Original Skillrove workflow.

No reviews yetFeatured

See what you can make

Skip the interactive example

Loading the working example…

Open example
Interact with the sample. Scroll inside to explore.
Made with this SkillA reproducible fix and proof handoff

Fictional fixture · real failing/fixed/passing test executions

Open full example

What you get

A reproducible fix and proof handoff

What you can make

  • Capture a real reproduction command, bounded logs and watched source hashes.
  • Compare competing causes and a narrow correction against the same regression checks.
  • Deliver an interactive before/after case and a reproducible fix handoff.
Explore delivered files5 groups
  • report/index.html
  • report/case.json
  • report/fix-handoff.md
  • report/delivery-manifest.json
  • private evidence/before.json + after.json

Good fit

Developers investigating a concrete local defect with a repeatable reproduction.

DeveloperFounder

Know the limits

Proof covers the captured reproduction; adjacent behavior still needs review.

About this skill

Capabilities: actual local debugging workflow with baseline capture, root-cause hypotheses, a narrow fix and fresh verification. Capture a concrete argv without a shell, bounded logs/timeouts, watched-source hashes and within-run changes. Compare eight evidence conditions, including identical reproduction commands and unchanged declared regression-test/fixture hashes.

Deliverables: interactive case/proof, hypothesis, terminal-evidence and source-scope views; case.json, Markdown fix handoff and delivery manifest. Original private capture files retain exact command outcomes and hashes. A one-command fictional demo executes a real failing test, fixes only a copied source and reruns the unchanged tests.

Read the full workflow

Requirements and run: Python 3.10+ standard library; the target project can require its own runtime. python3 scripts/run-example.py --output ./sample-run. For actual projects follow SKILL.md and references/first-run.md, capture before/fix/after and inspect the real diff.

Validation: actual numeric-sort fixture failed before and passed after the source correction. Independent bulk-pricing boundary defect fixed with preserved tests and 64 adjacent combinations verified. Incomplete/truncated logs, changed commands, changed reproduction inputs and leftover owned processes stay outside clean passing proof. UI, export hashes and mobile layout tested.

Limits: passing evidence covers the selected reproduction, not the whole application or production environment. Inputs are locally recorded evidence, not independent attestation; supplied root cause still needs code review. Finite-check capture terminates owned descendants on macOS/Linux; Windows can require manual descendant cleanup. Review paths/logs before sharing.

Version 1.0.0: original MIT-licensed tools, methodology attribution included. Individual purchase and active membership unlock the package.

Read skill facts →Choose a skill by task →
Debugging

FROM PACKAGE TO FIRST RESULT

Your first verified fix.

Run a real failing test, apply one isolated source correction and repeat the unchanged tests. The sample product is fictional; its commands and before/after evidence are actual executions.

View example result
US$4.99 · MIT licensePython 3.10+ · local project checksNo API key or paid provider
  1. 01

    Choose your agent

    These steps use a project folder. Open the project in your agent and run terminal commands from its root.

    Codex directory guide

    Package folder .agents/skills/debug-evidence-kit

  2. 02

    Add the package

    Download and unzip the package into the folder above. Keep all its files together, with SKILL.md directly inside that folder.

    Downloads require a verified account and purchase access. Choose a new folder if this Skill is already installed.

  3. 03

    Run the included sample

    Manual sample commands and technical details

    Read SKILL.md, review the generator and check its runtime requirements before running:

    cd .agents/skills/debug-evidence-kit
python3 scripts/run-example.py --output ./sample-run
    Your first result

    Open sample-run/report/index.html and inspect eight proof comparisons, hypotheses, before/after logs and source hashes. The fixture fails before the fix and passes after it while the regression-test hash stays identical. Follow references/first-run.md for the manual investigation.

    Python 3.10+ standard library only. Demo modifies only a fresh copied fixture. For user projects, capture-check.py executes your explicitly supplied finite local command. Passing proof covers that reproduction; inspect the diff and run adjacent checks before release.

    If sample-run already exists, change the output to ./sample-run-2. Existing files are never overwritten.

    Prefer to ask your agent?

    Use Debug Evidence Kit at .agents/skills/debug-evidence-kit. Read SKILL.md and review scripts/run-example.py. Run examples/case.json into a new output folder and show me all generated files. Keep fictional sample content clearly labeled.

Ready to use your own facts?

Copy examples/case.json to a new input file, replace all fictional facts, and generate into a new folder. Review the content before using or publishing it; the generator does not independently verify claims.

Reviews0

Share your first result

Tried this workflow? Help others decide with an honest review.

Before you choose this Skill

How do I get Debug Evidence Kit?

The listed individual package price is USD 4.99. Premium downloads require a confirmed purchase or active membership. Your agent subscription, model access and external provider credits are separate.

Compare access and pricing
Which agents can use this skill?

This release lists Codex, Claude Code, Cursor as intended installation hosts. Check the package’s dependencies and permissions before running it. These labels do not certify that every workflow has been tested in every agent.

Codex setupClaude Code setupCursor setup
What files can this workflow generate?

The documented 1.0.0 workflow lists: report/index.html; report/case.json; report/fix-handoff.md; report/delivery-manifest.json; private evidence/before.json + after.json. Some files or exports are optional; follow the package’s setup and inspect the delivered result.

Review all deliverablesInspect the working example
What should I check before using the result?

Python 3.10+ standard library only. Demo modifies only a fresh copied fixture. For user projects, capture-check.py executes your explicitly supplied finite local command. Passing proof covers that reproduction; inspect the diff and run adjacent checks before release.

Read the installation guideReview third-party notices

More skills to explore

View all skills
$4.99One-time purchase