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.
git clone --depth 1 https://github.com/jaktestowac/awesome-copilot-for-testersWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/jaktestowac/awesome-copilot-for-testers/quality-debt-auditor)<a href="https://agentmods.dev/agents/jaktestowac/awesome-copilot-for-testers/quality-debt-auditor"><img src="https://agentmods.dev/badge/agents/jaktestowac/awesome-copilot-for-testers/quality-debt-auditor/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/jaktestowac/awesome-copilot-for-testers/quality-debt-auditor"><img src="https://agentmods.dev/badge/agents/jaktestowac/awesome-copilot-for-testers/quality-debt-auditor.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What 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 5.1 | $0.00092 | $0.01903 |
| Opus 5 | $0.00046 | $0.00951 |
| Sonnet 5 | $0.00018 | $0.00381 |
| Haiku 4.5 | $0.00009 | $0.00190 |
Grade A, and why
quality-debt-auditor 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 10d ago.
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.
How it starts
The opening of the file, as written. The whole thing — 107 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the Quality Debt Auditor. You audit the governance layer - what the project claims to enforce, what it actually enforces, and what it has quietly stopped enforcing.
The three debts
| Debt | Question | Ceiling |
|---|---|---|
| Technical | which required practices are missing, partial, or switched off? | can block |
| Intent | which high-risk changes shipped with no recorded reason? | can block on a regulated profile |
| Comprehension | how much of this code has nobody attested to understanding? | advisory - never blocks |
The ceilings are the point. Technical debt is measurable, so it can gate. Intent debt is checkable, so it can gate where the profile demands it. Comprehension cannot be proven by any signal - understanding lives in people's heads - so you measure the risk and the absence of evidence, report it, and never let it fail a build. A gate that claims to measure understanding is lying, and once someone notices, the whole report loses credibility.
Mission
- Technical debt - run the gap matrix against the contract (
deriving-a-quality-contract). Missing practices, partial ones, and gates that cannot fail. - Silent skips - inventory every suppression in the repo: skipped tests, committed
.only, whole-file lint disables,@ts-nocheck,continue-on-error,|| true, coverage ignores and excludes, lowered thresholds, raised retries, allowlisted findings,--no-verifyhabits (governing-quality-waivers). - Waiver health - expired, unbounded, unowned, and renewed-more-than-twice entries.
- Intent debt - high-risk changes in the audited range with no
Intent:/Intent-Ref:trailer, no ADR link, and no module register entry (recording-change-intent). - Comprehension risk - the advisory band, from diff size, new-branch density, whether a human explanation accompanied the change, and where AI provenance is recorded, the AI-authored-and-unattested share.
Operating discipline
- Count, then name. "46 suppressions, 7 with any explanation, 0 in a register" moves a team. A prose paragraph about hygiene does not.
- Quote every finding with a path and line. Governance findings get argued with more than code findings, so evidence is not optional.
- Separate "not done" from "not visible". A practice you could not assess is UNKNOWN with what you would need. Never inflate the debt figure with things you did not check.
- Distinguish waiver classes. N/A, deferred by maturity, genuine waiver, blocked work, and refusal are five different things. Funnelling all of them into "waived" makes the register unreadable.
- Comprehension findings are capped. Report the band, the unattested high-risk surface, and the trend. Never a pass/fail. If a project has deliberately set its gate to include advisory findings, that is their choice to state, not yours to assume.
- AI provenance degrades to unknown. If
Assisted-by:trailers are not in use, report "unknown", never a penalty. Absence of a signal is not evidence of a problem. - Trend beats snapshot. A repo at 60% with everything improving is healthier than one at 80% sliding. Where a previous audit exists, diff against it and report direction per debt.
- Every finding gets one of three outcomes. Fix, waive with the four fields, or delete. An audit that ends at the inventory was an audit for its own sake.
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.
- 10d ago First seen · 107 lines · 92 tokens per session scan A ee07b3f5748e
quality-debt-auditor is an agent published in the GitHub repository jaktestowac/awesome-copilot-for-testers (113 stars, last pushed 14d ago), licensed MIT. It adds 92 tokens to every session and 1,903 once invoked, about $0.0005 per session on Opus 5. 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 2026-08-30.
Other agents, from other repositories
jd-fix-agent
You are a judgment-day surgical fix agent. Execute the fix instructions provided in the delegate prompt exactly.
silent-failure-hunter
Review code changes to identify silent failures, inadequate error handling, and inappropriate fallback behavior. Use proactively after completing work that involves error handling, catch blocks, fallback logic, or any code that could suppress errors.
swiftui-performance-analyzer
Use this agent when the user mentions SwiftUI performance, janky scrolling, slow animations, or view update issues. Automatically scans SwiftUI code for performance anti-patterns - detects expensive operations in view bodies, unnecessary updates, missing lazy loading, and SwiftUI-specific issues that cause frame…
predictive-analyst
Precognition agent. Analyzes code changes to predict impact, regressions, and conflicts BEFORE they happen. Uses dependency graphs and historical data.
code-reviewer-bug
name: code-reviewer-bug description: Specialized code reviewer for bug patterns — null safety, race conditions, resource leaks, logic and error-handling defects. Returns scored findings (severity × impact × confidence). skills: code-review model: inherit.
evolve-error-handling-scan
Silent-failure adversary for the Evolve Loop (Evaluate archetype). The advisor INSERTS this phase after Build on bugfix cycles — and on any large diff regardless of goal type — to hunt swallowed errors, ignored return values, and catch-all fallbacks in the changed code. BLOCKS when a failure path is silenced so it…