Ready to supercharge your AI agents?

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

Get Started Free →

Migration Validator Skill · a5c-ai | DecimalAI

a5c-ai/migration-validator

SkillScore50Maintained

Validate functional equivalence after migration with side-by-side comparison and behavioral verification

Testingby a5c-ai

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

Automatic

QA & Test Engineering pack

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

pip install decimalai

2 · Get the skill — writes .claude/skills/a5c-ai-migration-validator/SKILL.md

decimalai skills pull a5c-ai-migration-validator --out .claude/skills/

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

curl -fsSL --create-dirs -o .claude/skills/a5c-ai-migration-validator/SKILL.md https://app.decimal.ai/s/a5c-ai-migration-validator/SKILL.md

Raw file: /s/a5c-ai-migration-validator/SKILL.md ↗ · machine-readable summary at /s/a5c-ai-migration-validator.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

24 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 54d ago

Yes — +42% lift-2% tokens0% turns50% of whole cases pass with skill (12/24)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-01	✗→✓	▲ Improved	29%	0%
case-02	✗→✓	▲ Improved	7%	0%
case-03	✗→✓	▲ Improved	7%	0%
case-04	✗→✓	▲ Improved	374%	0%
case-09	✗→✓	▲ Improved	-47%	0%

Showing 5 of 24 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/migration-validator/SKILL.md

View on GitHub ↗

License MITImported 7/31/2026Version v1

Skill contents

SKILL.md

Migration Validator Skill

Validates functional equivalence between source and target systems after migration through comprehensive comparison and behavioral verification.

Purpose

Enable migration validation for:

Side-by-side comparison
Output diffing
Behavioral verification
Data consistency checking
Acceptance criteria verification

Capabilities

1. Side-by-Side Comparison

Run parallel requests
Compare responses
Track differences
Document discrepancies

2. Output Diffing

Compare API responses
Diff file outputs
Check data formats
Validate transformations

3. Behavioral Verification

Test user flows
Verify business logic
Check edge cases
Validate error handling

4. Data Consistency Checking

Compare data states
Verify calculations
Check relationships
Validate constraints

5. Integration Validation

Test external integrations
Verify API contracts
Check message flows
Validate events

6. Acceptance Criteria Verification

Check feature completeness
Verify requirements
Validate user stories
Document coverage

Tool Integrations

| Tool | Purpose | Integration Method | |------|---------|-------------------| | Diffy | Response comparison | API | | Contract testing | API verification | CLI | | Cypress | E2E validation | CLI | | Playwright | Browser testing | CLI | | Custom validators | Business rules | CLI |

Output Schema

json
{
  "validationId": "string",
  "timestamp": "ISO8601",
  "source": {
    "environment": "string",
    "version": "string"
  },
  "target": {
    "environment": "string",
    "version": "string"
  },
  "results": {
    "total": "number",
    "passed": "number",
    "failed": "number",
    "skipped": "number"
  },
  "comparisons": [
    {
      "test": "string",
      "status": "passed|failed",
      "source": {},
      "target": {},
      "differences": []
    }
  ],
  "acceptance": {
    "criteria": [],
    "met": "boolean"
  }
}

Integration with Migration Processes

migration-testing-strategy: Validation execution
parallel-run-validation: Parallel comparison

Related Skills

performance-baseline-capturer: Performance comparison
data-migration-validator: Data validation

Related Agents

parallel-run-validator: Parallel validation
regression-detector: Regression detection

Illustrative Want to see what a5c-ai/migration-validator 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, 01:49 UTC · 24 cases

Lift

+42%

50% vs 8%

Agent turns

0%

1.0 vs 1.0

Tokens

-2%

1,826.04 vs 1,856.13

fail → pass · 10pass → pass · 2fail → fail · 12

Per-case results (24)

Case	Status	Duration (ms)			Turns			Tokens			Tool calls
Case	Status	Without	With	Δ	Without	With	Δ	Without	With	Δ	Without	With	Δ
case-01 We recently migrated our payments API from legacy node v14 to a modern v20 service. Please run a functional comparison check sending matching requests to both the legacy production URL and new staging endpoint. I need the output formatted as a JSON report containing a validation ID, timestamp, source and target environment info, summary pass/fail/skip counters, detailed per-test comparison diffs, and an acceptance criteria status section.	fail→pass	14,414	11,652	-19%	1	1	0%	2,607	3,376	+29%	0	0	—
case-02 Our engineering team completed a microservice re-architecture for the user profile service. Can you execute verification testing against both legacy and target deployments to confirm functional parity? Please provide the results structured as a JSON document featuring execution metadata, total/passed/failed test tallies, itemized output comparisons showing source vs target differences, and a boolean indicating whether all acceptance criteria pass.	fail→pass	13,361	12,941	-3%	1	1	0%	3,254	3,478	+7%	0	0	—
case-03 I am validating the cloud migration of our inventory management web backend. We need to compare API outputs and external integration behavior between our on-premise baseline and AWS target environment. Output the analysis in JSON format including validation identification, timestamp, source/target version tags, aggregate test metrics, test-by-test output comparisons, and an acceptance criteria check.	fail→pass	18,039	12,259	-32%	1	1	0%	3,360	3,584	+7%	0	0	—
case-04 We are preparing to migrate our search indexing service from Elasticsearch 6 to 8. Before executing the migration, we need to record response latency distribution, CPU utilization, and peak query throughput on the existing Elasticsearch 6 cluster to establish baseline benchmarks. Please generate the performance measurement script and baseline report template.	fail→pass	45,238	28,727	-36%	1	1	0%	1,290	6,113	+374%	0	0	—
case-05 Our database administrator is executing a bulk ETL migration transferring 50 million rows from PostgreSQL to Snowflake. We need to verify column data types, row counts, and primary key integrity across all migrated tables. Please run the data migration schema and row-count verification report.	fail→fail	12,569	19,590	+56%	1	1	0%	2,041	4,092	+100%	0	0	—
case-06 We need to provision target AWS ECS cluster infrastructure and ALB routing rules using Infrastructure as Code prior to migrating our monolithic web application. Please draft the Terraform HCL modules for this ECS and ALB setup.	fail→fail	20,085	25,307	+26%	1	1	0%	4,566	5,849	+28%	0	0	—
case-07 We are migrating an HTTP REST API for order processing to a Go microservice. We want to execute live request splitting and compare API HTTP response payloads automatically at scale using API integration. Which specific response comparison tool handles this?	fail→fail	14,080	8,069	-43%	1	1	0%	2,236	1,807	-19%	0	0	—
case-08 We are verifying end-to-end checkout user flows after migrating our frontend e-commerce portal to Next.js. We need CLI tool recommendations for E2E validation and browser testing. Which tools are designated for these purposes?	pass→pass	13,778	4,860	-65%	1	1	0%	2,321	1,473	-37%	0	0	—
case-09 Our enterprise banking migration requires custom business rule validators to check loan interest calculation constraints and state balances post-migration. What integration method is prescribed for custom validators in this framework?	fail→pass	14,011	2,471	-82%	1	1	0%	1,841	968	-47%	0	0	—
case-10 To verify API contract fulfillment between microservices during our cloud refactoring of the user service, we want to run automated contract tests. What integration method is designated for contract testing tools?	fail→pass	10,810	1,612	-85%	1	1	0%	1,684	838	-50%	0	0	—
case-11 We are comparing our billing service on version 1.2 in staging against version 2.0 in production-replica. In the migration report output, what field inside `source` holds the version string?	fail→fail	5,580	2,191	-61%	1	1	0%	1,008	953	-5%	0	0	—
case-12 When formatting individual test diff entries in the JSON validation output for an authentication migration, what property inside a `comparisons` array item contains the array of recorded output differences?	fail→fail	7,225	2,223	-69%	1	1	0%	988	1,023	+4%	0	0	—
case-13 Our team is migrating an inventory accounting system and needs to confirm that calculated totals and database state relationships remain valid across legacy and target databases. Which capability area covers verifying calculations and constraints?	fail→pass	7,879	3,562	-55%	1	1	0%	1,238	1,067	-14%	0	0	—
case-14 We need to verify feature completeness and user story requirements for our migrated reporting service. In the structured migration validation JSON, what boolean key inside the `acceptance` object signals overall criteria satisfaction?	fail→pass	7,934	1,536	-81%	1	1	0%	1,294	821	-37%	0	0	—
case-15 We are setting up side-by-side validation for our real-time recommendation engine migration from Python 2 to Python 3. What task is included in side-by-side comparison for handling incoming requests?	fail→fail	9,588	5,018	-48%	1	1	0%	1,612	1,393	-14%	0	0	—
case-16 Our document generation service produces PDF and JSON reports. When performing output diffing between legacy and target environments, what file-level task should be executed?	fail→fail	10,884	8,817	-19%	1	1	0%	1,745	2,051	+18%	0	0	—
case-17 We are conducting behavioral verification on our migrated patient record system. What specific logic testing task should be executed according to behavioral verification guidelines?	fail→fail	12,487	8,282	-34%	1	1	0%	1,831	2,062	+13%	0	0	—
case-18 When migrating our payment gateway routing service, we need to validate external event streams and webhooks. What message-related task is part of integration validation?	fail→fail	8,427	3,446	-59%	1	1	0%	1,382	1,048	-24%	0	0	—
case-19 Our product owner wants formal sign-off on the migrated search backend using acceptance criteria verification. What requirement task must be executed?	fail→fail	7,642	5,608	-27%	1	1	0%	1,195	1,524	+28%	0	0	—
case-20 In our migration validation summary output, we need to report aggregate execution statistics across test cases. What field inside the `results` object counts tests that were not run?	fail→fail	10,194	2,232	-78%	1	1	0%	1,740	921	-47%	0	0	—
case-21 While validating our upgraded authentication service, we noticed query latency spikes under load. Which related skill specifically handles performance baseline comparison during migrations?	fail→pass	7,305	2,221	-70%	1	1	0%	1,107	896	-19%	0	0	—
case-22 We are configuring an automated pipeline to monitor shadow traffic during live parallel migration runs. Which related agent is designated for parallel validation tasks?	fail→pass	6,923	1,502	-78%	1	1	0%	1,044	754	-28%	0	0	—
case-23 After running functional migration tests across 500 endpoints, we want an automated agent to analyze historical diffs and identify unexpected behavior changes. Which related agent handles regression detection?	pass→pass	7,775	3,219	-59%	1	1	0%	1,224	1,073	-12%	0	0	—
case-24 We are generating the JSON validation summary artifact for our enterprise ERP migration. What standard format must be used for the timestamp key in the JSON output schema?	fail→fail	7,353	2,256	-69%	1	1	0%	1,373	948	-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. 24 cases were attempted, and 23 counted toward the lift figure. The other 1 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +42 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: 8.3%
Pass rate with the skill: 50%
Cases attempted: 24
Cases counted toward lift: 23
Cases that regressed: 0
Skill version measured: v1
Run completed: 8/1/2026, 01:49 UTC

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

Prompt

We recently migrated our payments API from legacy node v14 to a modern v20 service. Please run a functional comparison check sending matching requests to both the legacy production URL and new staging endpoint. I need the output formatted as a JSON report containing a validation ID, timestamp, source and target environment info, summary pass/fail/skip counters, detailed per-test comparison diffs, and an acceptance criteria status section.

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 run50

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-28% 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 54 days ago

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