# Speck Debug

> Diagnoses errors and failed Speck workflows systematically. Use when tests, tools, or expected behavior break.

- Skill: `telum-ai/speck-debug` (Agent Skill, multi-file: 5 files)
- Install (CLI): `npx skillmds@latest add telum-ai/speck-debug`
- Raw SKILL.md: https://api.skillmd.com/api/skills/telum-ai/speck-debug/raw
- Safety review: pending (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: telum-ai (https://skillmd.com/u/telum-ai)
- Updated: 2026-09-22
- Page: https://skillmd.com/skills/telum-ai/speck-debug

---


# speck-debug

Cheap keys: debug phase (triage → hypothesize → evidence → fix). Do not Read later phases early.

1. MUST Read `references/triage.md` first (classify failure).
2. After symptoms classified: MUST Read `references/hypotheses.md`.
3. Before claiming root cause: MUST Read `references/evidence.md` (repro required).
4. Only after root cause is evidenced: MUST Read `references/fix-loop.md`; fix + verify.