Agentic Tech Debt — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited Agentic Tech Debt (Plugin) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Catches AI-introduced tech debt at write-time
Works with [any coding agent](#install). Any stack.
Two decades of tech-debt research, distilled into a plugin and validated across dozens of codebases.
<img src="./demo/debt-ops.gif" width="720" alt="debt-ops in Claude Code: the agent edits api/checkout.ts and casts a value to as any to clear a type error; debt-ops catches the loosened type at write-time and logs +1 entry: as-any-checkout-payload (A)" />
Just needs a git repo and Python 3.10+.
<a id="claude-code"></a> <img src="./assets/agents/claude.svg" height="16" alt="" align="absmiddle" /> Claude Code
/plugin marketplace add bcanfield/agentic-tech-debt
/plugin install debt-ops@agentic-tech-debt<details> <summary><img src="./assets/agents/codex.svg" height="16" alt="" align="absmiddle" /> <b>Codex</b></summary>
codex plugin marketplace add bcanfield/agentic-tech-debt
# then, inside Codex: /plugins → install debt-ops</details>
<details> <summary><img src="./assets/agents/cursor.svg" height="16" alt="" align="absmiddle" /> <b>Cursor</b></summary>
One-click install via the plugin (hooks + skills bundled):
/add-plugin # then pick debt-ops, or browse cursor.com/marketplacePrefer to install by hand? Copy hooks.local.json + the scripts into .cursor/ and the skills into .agents/skills/:
mkdir -p .cursor/hooks .agents/skills
cp cursor/hooks/hooks.local.json .cursor/hooks.json
cp cursor/hooks/*.py .cursor/hooks/
cp -r cursor/skills/debt-ops-* .agents/skills/See the Cursor adapter README for details.
</details>
<details> <summary><img src="./assets/agents/copilot.svg" height="16" alt="" align="absmiddle" /> <b>GitHub Copilot CLI</b></summary>
copilot plugin marketplace add bcanfield/agentic-tech-debt
copilot plugin install debt-ops@agentic-tech-debtThen run debt-ops-init once — Copilot has no per-session inject, so this writes the disciplines into your charter (.github/copilot-instructions.md or AGENTS.md). See the Copilot adapter README for the manual-install alternative and details.
</details>
<details> <summary><img src="./assets/agents/gemini.svg" height="16" alt="" align="absmiddle" /> <b>Any other agent</b> (Gemini CLI, Windsurf, opencode…)</summary>
Via portable Agent Skills:
npx skills add bcanfield/agentic-tech-debtThen run debt-ops-init once to write the disciplines into your AGENTS.md.
</details>
<details> <summary>Commands</summary>
Claude Code /debt-ops:<name> · Codex $<name>:
drop A, drop A,C, or drop all.CLAUDE.md/AGENTS.md so the team shares them.</details>
AI writes more code and cleans up less of it: across 211M lines, one study saw refactoring more than halve while duplicated code kept climbing. So debt piles up faster than you can review it. The fixes are well understood: keep it visible, pay it down over time, record the decisions. debt-ops wires those into the agent loop as nine cited pillars. Full research synthesis.
<img src="./demo/concept/debt-ops-concept.gif" width="720" alt="A as any cast in api/checkout.ts is caught at write-time and files itself as a registry entry +1 entry: as-any-checkout-payload (A); a // TODO nit is caught the same way; the nit is pruned and the kept entry is paid down (strikethrough + green check); the registry empties to zero; debt-ops wordmark." />
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.