Skip to content
Skills Library
Programming · Analysis

decant-debug

/decant-debug

Diagnose a failure from reproducible evidence and isolate its root cause. Use for regressions, flaky behavior, stuck automation, or failed checks; edit only when the user also asked for a fix.

Repository
decant ↗
Licence
MIT
Last repository change
25.10.2026

Assessment and security

100/ 100Class: good
Format25 / 25
Frontmatter with name and description, valid name, description up to 1,024 characters, name matching the folder.
Discoverability25 / 25
Description explains when to use the skill, has a reasonable length and uses third person.
Content35 / 35
At least 150 words and at most 500 lines, headings, step lists, examples or code.
Hygiene15 / 15
No placeholders such as TODO, no Windows paths, with a licence.
Riskno signals detected in the scan
Files in this package
2
Scripts
0
HTML comments
0
URLs in SKILL.md
0
Deprecation

No signals detected. Repository changed on 25.10.2026, without references to retired models or APIs.

Licence

MIT, open: use and modify while retaining author attribution.

What it does and when to use it

Diagnose a failure from reproducible evidence and isolate its root cause. Use for regressions, flaky behavior, stuck automation, or failed checks; edit only when the user also asked for a fix.

Original author description

Diagnose a failure from reproducible evidence and isolate its root cause. Use for regressions, flaky behavior, stuck automation, or failed checks; edit only when the user also asked for a fix.

How it works

  1. Diagnostic loop
  2. Authorization boundary
  3. Output contract
Original SKILL.md headings · 33 lines, 192 words
SKILL.md preview
---
name: decant-debug
description: Diagnose a failure from reproducible evidence and isolate its root cause. Use for regressions, flaky behavior, stuck automation, or failed checks; edit only when the user also asked for a fix.
---

# Decant Debug

Prefer one falsifiable hypothesis over a sequence of speculative patches.

## Diagnostic loop

1. Capture the exact symptom, environment, expected behavior, and first known failure.
2. Reproduce with the smallest safe command or fixture.
3. Locate the failing boundary: input, configuration, process, network, storage, adapter, or output.
4. Collect evidence on both sides of that boundary.
5. State one hypothesis and the observation that would falsify it.
6. Run the smallest experiment that distinguishes competing causes.
7. If a fix is authorized, apply the minimal repair and add a regression check.

After three materially similar failed repair attempts, stop patching. Revisit the reproduction, boundary, and assumptions.

## Authorization boundary

A request to diagnose does not authorize code changes. A request to fix does not automatically authorize data deletion, credential changes, deployment, or publication. Use backup-first procedures for stateful repair.

## Output contract

Return:

- reproduced symptom and command;
- root cause with evidence, or ranked remaining hypotheses;
- affected and unaffected boundaries;
- fix and regression test, if authorized;
- operational risks and rollback.

Distinguish a health label such as `connected` or `configured` from proof that useful work completed.
Files in this package (2)
  • SKILL.md 2 KB
  • agents/openai.yaml 1 KB

Versions

The database has no other skills named decant-debug.

Database snapshot: 13.11.2026 · package version 47fc015a57dc (SHA-256 digest of the entire package)

Downloads in the last 30 days: 0 · ranking

Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.

Similar skills

  • 100oma-debug
    Diagnose a reproducible failure, fix its cause, and verify the regression. Use for crashes, incorrect behavior, and failing tests.
  • 100oma-debug
    Diagnose a reproducible failure, fix its cause, and verify the regression. Use for crashes, incorrect behavior, and failing tests.
  • 100diagnose
    Disciplined debugging loop for broken behaviour, failing builds, flaky commands, hung processes or regressions. Use when asked to debug or find a root cause.
  • 86debug
    Use when something fails or behaves wrong and the cause is not obvious - an error, traceback, failing test, crash, hang or wrong output. Systematic reproduce, isolate, fix, verify.
  • 100duck-debug
    Socratic debugging: root-cause questioning or read-only evidence tracing. Use when: "debug this", "why is X broken", "trace this failure", "where is this used", "map callers".
  • 100duck-debug
    Socratic debugging: root-cause questioning or read-only evidence tracing. Use when: "debug this", "why is X broken", "trace this failure", "where is this used", "map callers".

Feedback and reports

Your message goes to Bartosz Sokoliński. An email address is needed only for a reply. See the privacy policy.