Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Get Started Free →

Technical Debt Quantifier Skill · a5c-ai | DecimalAI

a5c-ai/technical-debt-quantifier

SkillScore43Maintained

Measure, categorize, and prioritize technical debt for migration planning and remediation

Code Reviewby a5c-ai

Links this skill to your agents and keeps it current with the author’s updates.

Automatic

Coding pack

1 · One-time — install the CLI (docs ↗)

pip install decimalai

2 · Get the skill — writes .claude/skills/a5c-ai-technical-debt-quantifier/SKILL.md

decimalai skills pull a5c-ai-technical-debt-quantifier --out .claude/skills/

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/a5c-ai-technical-debt-quantifier/SKILL.md https://app.decimal.ai/s/a5c-ai-technical-debt-quantifier/SKILL.md

Raw file: /s/a5c-ai-technical-debt-quantifier/SKILL.md ↗ · machine-readable summary at /s/a5c-ai-technical-debt-quantifier.json ↗

In productionGlobal

how this skill behaves on real traffic · last 30 days

No production data for this skill yet — usage numbers appear here once agents start running it.

Anonymized aggregate across every org using this skill — proof from real production use, not a demo.

Does this skill help? · Benchmark: with vs. withoutsign in to run this on your agents

23 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 56d ago

Yes — +35% lift+9% tokens0% turns43% of whole cases pass with skill (10/23)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-01	✗→✓	▲ Improved	-8%	0%
case-02	✗→✓	▲ Improved	-10%	0%
case-04	✗→✓	▲ Improved	26%	0%
case-08	✗→✓	▲ Improved	32%	0%
case-09	✗→✓	▲ Improved	-21%	0%

Showing 5 of 23 cases

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 7/31/2026

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

a5c-ai / babysitter

library/specializations/code-migration-modernization/skills/technical-debt-quantifier/SKILL.md

View on GitHub ↗

License MITImported 7/31/2026Version v1

Skill contents

SKILL.md

Technical Debt Quantifier Skill

Measures, categorizes, and prioritizes technical debt to support informed decision-making for migration planning and debt remediation strategies.

Purpose

Enable technical debt management for:

Debt categorization and inventory
Remediation effort estimation
Interest calculation (ongoing cost)
Priority scoring
Trend tracking

Capabilities

1. Debt Categorization

Code debt (smells, complexity)
Architecture debt (coupling, cohesion)
Test debt (coverage gaps)
Documentation debt
Infrastructure debt

2. Remediation Effort Estimation

Estimate fix time per item
Calculate total remediation cost
Identify quick wins
Plan sprint allocation

3. Interest Calculation

Calculate ongoing maintenance cost
Estimate productivity impact
Project future debt growth
Model compound interest

4. Priority Scoring

Score by business impact
Weight by risk level
Factor in remediation cost
Calculate ROI of fixes

5. Debt-to-Value Ratio

Compare debt to feature velocity
Benchmark against industry
Track debt percentage
Set organizational targets

6. Trend Tracking

Monitor debt over time
Track remediation progress
Identify debt sources
Report debt velocity

Tool Integrations

| Tool | Purpose | Integration Method | |------|---------|-------------------| | SonarQube | Debt calculation | API | | CodeScene | Hotspot analysis | API | | Codacy | Quality metrics | API | | Code Climate | Maintainability | API | | NDepend | .NET debt analysis | CLI |

Output Schema

json
{
  "analysisId": "string",
  "timestamp": "ISO8601",
  "debt": {
    "total": {
      "estimatedHours": "number",
      "monetaryValue": "number",
      "items": "number"
    },
    "byCategory": {
      "code": {},
      "architecture": {},
      "test": {},
      "documentation": {}
    },
    "byPriority": {
      "critical": [],
      "high": [],
      "medium": [],
      "low": []
    }
  },
  "metrics": {
    "debtRatio": "number",
    "debtPerLoc": "number",
    "interestRate": "number"
  },
  "trends": {
    "thirtyDay": "number",
    "ninetyDay": "number"
  },
  "recommendations": []
}

Integration with Migration Processes

legacy-codebase-assessment: Debt quantification
technical-debt-remediation: Prioritization

Related Skills

code-smell-detector: Debt identification
static-code-analyzer: Quality metrics

Related Agents

technical-debt-auditor: Deep debt analysis

Illustrative Want to see what a5c-ai/technical-debt-quantifier would put in front of your model? Preview the rendered body and check its trigger phrases — no LLM call, no signup.

Latest benchmark · publisher runVerified

8/1/2026, 01:55 UTC · 23 cases

Lift

+35%

43% vs 9%

Agent turns

0%

1.0 vs 1.0

Tokens

+9%

2,441.96 vs 2,248.83

fail → pass · 8pass → pass · 2fail → fail · 13

Per-case results (23)

Case	Status	Duration (ms)			Turns			Tokens			Tool calls
Case	Status	Without	With	Δ	Without	With	Δ	Without	With	Δ	Without	With	Δ
case-01 Please analyze our backend repository and generate a complete technical debt assessment structured in JSON. The resulting data structure must include an analysis identifier, timestamp, and top-level debt summary containing total fix hours, estimated financial cost, and item count. Additionally, detail the debt split across categories (such as code, architectural, testing, and docs) and organize issues into critical, high, medium, and low priority arrays. Make sure to capture key metrics like debt ratio and interest rate, 30 and 90 day velocity trends, and a list of recommended remediation steps.	fail→pass	21,261	13,039	-39%	1	1	0%	4,131	3,805	-8%	0	0	—
case-02 We are preparing for an upcoming migration sprint and need a structured technical debt report for our core service. Deliver a JSON document that records an analysis ID, ISO timestamp, and total debt figures including hours required, dollar value, and total count. Group the findings by category and priority buckets from critical down to low, alongside metrics like debt per line of code and interest rate. Finally, include thirty and ninety day historical trend values along with strategic fix recommendations.	fail→pass	20,018	15,479	-23%	1	1	0%	3,831	3,444	-10%	0	0	—
case-03 Run a technical debt evaluation on our codebase to help us plan our refactoring roadmap. I need the response formatted as JSON with an analysisId, timestamp, and a debt object that breaks down total estimated hours, dollar impact, and item count. Categorize the findings into code, architecture, test, and documentation, and group them by priority arrays. Ensure the payload includes metrics like debtRatio and debtPerLoc, short-term and long-term trend data (30-day and 90-day), and an array of actionable recommendations.	fail→fail	18,060	22,453	+24%	1	1	0%	3,979	4,358	+10%	0	0	—
case-04 We are auditing a payment service written in Go. Output a JSON technical debt evaluation payload for this repository. The user requested category-level breakdown of the debt. Base models often put 'byCategory' as a top-level object or use keys like 'architecture_debt' or 'testing'. How should the category mapping be formatted in the output payload?	fail→pass	13,139	9,990	-24%	1	1	0%	2,169	2,735	+26%	0	0	—
case-05 When collecting technical debt metrics from NDepend for a .NET framework solution, what integration mechanism should be configured in the automation pipeline? Pipeline developers often default to REST API endpoints assuming standard web service access.	fail→fail	11,898	8,905	-25%	1	1	0%	1,989	2,068	+4%	0	0	—
case-06 Our CI pipeline aggregates static analysis from SonarQube, CodeScene, Codacy, and Code Climate. What interface type should be configured to query technical debt figures from these four services?	fail→fail	6,684	6,239	-7%	1	1	0%	1,098	1,829	+67%	0	0	—
case-07 During sprint planning, our team wants to model the financial impact of postponing technical debt cleanup for our inventory management module. What specific ongoing cost metric should be calculated to represent the delayed refactoring cost?	fail→fail	10,543	10,920	+4%	1	1	0%	1,720	2,438	+42%	0	0	—
case-08 We need to formulate a priority score for refactoring candidates across our microservices. Should priority scoring rely strictly on risk level and business impact, or what additional economic factors must be factored in?	fail→pass	18,732	20,094	+7%	1	1	0%	2,766	3,638	+32%	0	0	—
case-09 When running technical debt quantification on a monolithic repository, which specialized skill should be invoked specifically to discover individual code smells during the initial discovery phase?	fail→pass	7,242	2,058	-72%	1	1	0%	1,229	971	-21%	0	0	—
case-10 For deep technical debt analysis requiring contextual investigation across legacy repositories, which specialized agent is designed to perform the deep debt audit?	pass→pass	7,387	2,506	-66%	1	1	0%	1,351	1,045	-23%	0	0	—
case-11 Our engineering VP wants to benchmark our repository's debt burden against feature delivery velocity. What specific comparison ratio should be tracked to measure technical debt against velocity?	fail→fail	13,301	9,636	-28%	1	1	0%	2,180	2,184	+0%	0	0	—
case-12 In an enterprise migration workflow, how does technical debt quantification integrate with the legacy codebase assessment process?	fail→fail	16,911	12,407	-27%	1	1	0%	2,581	2,553	-1%	0	0	—
case-13 How does the technical debt quantifier interact with the technical debt remediation process during sprint execution?	pass→pass	14,342	9,489	-34%	1	1	0%	2,264	2,360	+4%	0	0	—
case-14 An automated scanner reports that 40% of critical paths in our auth service lack unit tests. Under which technical debt category does this issue belong in the assessment schema?	fail→fail	7,658	3,576	-53%	1	1	0%	1,276	1,133	-11%	0	0	—
case-15 A repository scan reveals high class coupling and low module cohesion between billing and user profile modules. Under which category should this debt item be classified?	fail→fail	5,132	2,718	-47%	1	1	0%	838	1,076	+28%	0	0	—
case-16 When analyzing debt remediation options for an upcoming 2-week sprint, what specific subset of debt items should be flagged to maximize immediate impact with minimal effort?	fail→fail	14,288	11,629	-19%	1	1	0%	1,992	2,424	+22%	0	0	—
case-17 Provide the exact JSON object schema structure for the metrics property in a technical debt quantification payload.	fail→pass	13,975	6,769	-52%	1	1	0%	3,405	1,299	-62%	0	0	—
case-18 What exact historical time windows are tracked in the trends section of the technical debt JSON output schema?	fail→pass	6,794	2,025	-70%	1	1	0%	1,181	988	-16%	0	0	—
case-19 Our Kubernetes deployment manifests use outdated API versions and hardcoded secrets. Which debt category under the debt taxonomy covers container configuration and deployment script issues?	fail→fail	9,012	5,008	-44%	1	1	0%	1,430	1,576	+10%	0	0	—
case-20 We identified a code smell where var is used instead of const or let across 150 JavaScript files. Generate a Python script using jscodeshift or babel to automatically parse the AST and rewrite these variable declarations in place.	fail→fail	15,610	15,671	+0%	1	1	0%	3,077	3,882	+26%	0	0	—
case-21 Write a SonarQube Quality Profile XML definition that enforces a maximum cyclomatic complexity of 15 and requires at least 80% line coverage for Java projects.	fail→fail	11,425	13,158	+15%	1	1	0%	2,333	3,360	+44%	0	0	—
case-22 Here is a Python function for calculating discount rates that lacks test coverage. Write a complete pytest test suite covering all edge cases and branch paths for this function.	fail→fail	10,606	13,820	+30%	1	1	0%	2,299	3,576	+56%	0	0	—
case-23 We need to present a financial cost justification for refactoring our legacy monolith to the executive board. Generate a JSON payload adhering to the debt quantifier output schema containing monetary estimation for 500 hours of technical debt.	fail→pass	11,900	13,435	+13%	1	1	0%	2,604	3,423	+31%	0	0	—

These cases were also screened by the skill’s intent / safety pass.See the Trust review ↗Reproduce this benchmark in Colab ↗

How this skill was measured

DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 23 cases were attempted. The headline lift of +35 percentage points is the difference between those two pass rates over the 23 comparable cases.

Model measured: gemini-3.6-flash
Graded by: gemini-3.6-flash
Grading method: judged
Verification: verified
Pass rate without the skill: 8.7%
Pass rate with the skill: 43.5%
Cases attempted: 23
Cases counted toward lift: 23
Cases that regressed: 0
Skill version measured: v1
Run completed: 8/1/2026, 01:55 UTC

Measured example — same prompt, with vs. without the skill · from the benchmark run · case case-01

Prompt

Please analyze our backend repository and generate a complete technical debt assessment structured in JSON. The resulting data structure must include an analysis identifier, timestamp, and top-level debt summary containing total fix hours, estimated financial cost, and item count. Additionally, detail the debt split across categories (such as code, architectural, testing, and docs) and organize issues into critical, high, medium, and low priority arrays. Make sure to capture key metrics like debt ratio and interest rate, 30 and 90 day velocity trends, and a list of recommended remediation steps.

Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.

Show both answers

SkillScore blends 4 signals1 of 4 complete

BenchmarkPass rate WITH skill on a verified run43

Live evalNeeds ≥10 evaluated traces in 30 days

AI ratingNeeds ≥10 rated traces

AdoptionNeeds ≥10 serves across ≥2 orgs in 30 days

Efficiency · separate axis (at equal correctness)+0% turns-6% tokens

Shown for transparency; not credited unless a verified run shows no correctness regression.

Backed by 1 of 4 signals — each added signal makes the score more robust.

Verified on gemini-3.6-flash · re-tested 55 days ago

Other measured skills in the registry, with their headline benchmark lift.