Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/cbmono/ai-setup/techdebtgit clone --depth 1 https://github.com/cbmono/ai-setupWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00000 | $0.00517 |
| Opus 5 | $0.00000 | $0.00259 |
| Sonnet 5 | $0.00000 | $0.00103 |
| Haiku 4.5 | $0.00000 | $0.00052 |
Grade A, and why
techdebt scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
Find and remove duplicated logic, dead code, and low-value abstractions across the whole repo (use the built-in /simplify skill instead if you only want to clean up the current diff). Run this at the end of a session to keep tech debt from accumulating.
Steps
- Load the backlog. Read
.claude/techdebt.mdif it exists — that's the curated list of known tech debt the user previously chose to defer. Use it to skip re-reporting items already on the backlog. - Scan in parallel. Spawn subagents (or run Grep/Glob in parallel) across the main source tree, excluding
node_modules, build output, generated code, and vendored deps. If$ARGUMENTSnames a folder, scope the scan there. - Look for:
- Near-duplicate functions (same body, different names)
- Copy-pasted blocks of 6+ lines across files
- Dead exports: exported symbols with zero imports
- Wrapper functions that only forward to one other function
- Utility files reimplementing something already in
lodash/remeda/ node built-ins already present inpackage.json - Commented-out blocks older than a week (
git blameto confirm)
- Report findings as a ranked list (highest confidence first), each with file paths, line numbers, and the suggested action. Mark items already in the backlog so the user sees them in context.
- Ask per finding: fix now, defer, or reject.
- Apply approved fixes directly. Keep edits surgical — match surrounding style, preserve public API shapes, don't introduce new dependencies. If a fix would change behaviour (not just structure), flag it and skip.
- Update the backlog at
.claude/techdebt.md:- Append newly deferred findings.
- Remove items that were fixed or rejected this session.
- Dedupe: if a new finding matches an existing entry (same file + rough location + same category), keep the existing one, don't add a duplicate.
- One line per item:
- <category> — <file>:<line> — <one-line description>. No prose. - Create the file if it doesn't exist. Single top heading, nothing else — it's a rolling backlog, not a report.
The backlog holds accepted tech debt only. Don't log resolved items; don't let it grow indefinitely.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 25 lines · 0 tokens per session scan A 5269720415c7
techdebt is a command published in the GitHub repository cbmono/ai-setup (2 stars, last pushed 8d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 517 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.