Assess Technical Debt
Project-wide technical-debt scan ranked by impact vs refactor effort into a report
Test report
- Verdict
- Works with setup
- Score
- Tested
- Jul 21, 2026
- Environment
- Claude Code 2.x (agent harness)
- Upstream re-checked
- Aug 10, 2026 · 1846fa7
Fetched SKILL.md and references/debt-reviewer.md via raw GitHub; both parse and all body references resolve (debt-reviewer.md, sibling skills evaluate-findings and frontend-design all HTTP 200). Frontmatter is valid with name + an explicit trigger-rich description; no hardcoded machine paths (uses .turbo/ relative and $TMPDIR); no security smells (no curl|sh, base64, or exfiltration — worktree installs are scoped to the user's own project). I could not measure output because the skill's core is Step 2's concurrent model:opus Agent subagent fan-out plus Steps 3/5 invoking sibling Turbo skills — none executable by me as a spawned subagent in a clean env — so any hand-written report would skip the actual mechanism. Judged 5 trigger phrasings: activate = "assess the technical debt in this repo", "what should we refactor first", "where's the worst code rot in our codebase"; should-not = "refactor this function to reduce complexity" (edit, not analysis), "review my PR diff for bugs" (diff-scoped, skill itself points to /review-code) — all 5 judged correctly.
Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works
- Installs cleanly 5/5
- Triggers reliably 5/5
- Output vs. baseline 5/10
- Docs & honesty 5/5
What Assess Technical Debt does
Orchestrates a project-wide, analysis-only technical-debt assessment: fans out concurrent subagents over source partitions to find complexity hotspots, deprecated APIs, duplication, and architecture rot, then ranks findings by impact vs refactor effort into .turbo/technical-debt.md and a styled HTML report. Triggers on requests like "assess technical debt", "what should we refactor", or "where is the code rot". Part of the Turbo skill suite and depends on its sibling skills evaluate-findings and frontend-design.
How to install Assess Technical Debt
git clone --depth 1 https://github.com/tobihagemann/turbo.git /tmp/assess-technical-debt-src
mkdir -p ~/.claude/skills
cp -R /tmp/assess-technical-debt-src/claude/skills/assess-technical-debt ~/.claude/skills/assess-technical-debt
# Standalone-incomplete: this skill orchestrates sibling Turbo skills at runtime.
# For the full pipeline (Step 3 /evaluate-findings, Step 5 /frontend-design) install those too, or the whole suite:
# cp -R /tmp/assess-technical-debt-src/claude/skills/* ~/.claude/skills/
# Also relies on Turbo's TaskCreate/TaskList + Agent subagent fan-out; see claude/SETUP.md in the repo.
Skills live in ~/.claude/skills/ (global) or .claude/skills/
(per-project). Restart Claude Code after installing.
Commands — how to trigger Assess Technical Debt
-
/assess-technical-debtProject-wide technical-debt scan ranked by impact vs refactor effort into a report
It also activates on plain-language prompts like these:
-
Find our worst technical debt hotspots -
What should we refactor first in this codebase -
Assess technical debt across the whole project
Frequently asked questions
- Is the Assess Technical Debt skill free?
- Yes. The skill itself is free from tobihagemann/turbo. SkillProof publishes the install command and an independent test verdict at no cost.
- Does Assess Technical Debt work with Claude Code?
- We tested it with Claude Code 2.x (agent harness) on Jul 21, 2026. Verdict: Works with setup. Fetched SKILL.md and references/debt-reviewer.md via raw GitHub; both parse and all body references resolve (debt-reviewer.md, sibling skills evaluate-findings and frontend-design all HTTP 200). Frontmatter is valid with name + an explicit trigger-rich description; no hardcoded machine paths (uses .turbo/ relative and $TMPDIR); no security smells (no curl|sh, base64, or exfiltration — worktree installs are scoped to the user's own project). I could not measure output because the skill's core is Step 2's concurrent model:opus Agent subagent fan-out plus Steps 3/5 invoking sibling Turbo skills — none executable by me as a spawned subagent in a clean env — so any hand-written report would skip the actual mechanism. Judged 5 trigger phrasings: activate = "assess the technical debt in this repo", "what should we refactor first", "where's the worst code rot in our codebase"; should-not = "refactor this function to reduce complexity" (edit, not analysis), "review my PR diff for bugs" (diff-scoped, skill itself points to /review-code) — all 5 judged correctly.
- What is the Assess Technical Debt SkillProof Score?
- 8.0/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 5/10, docs & honesty 5/5.
- How do I install Assess Technical Debt?
- Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
- Can I use Assess Technical Debt with Cursor, Copilot, Gemini CLI, Codex or other AI tools?
- The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.