Build or modify backend APIs, services, jobs, persistence, and integrations while preserving contracts, authorization, data integrity, and failure behavior.
SkillScore73Maintained
Automatic
◆Coding pack
Last verified 9/1/2026
What it changes
Passed 12 of 17 test cases with the skill, and 11 of 17 without it.
Without skill64%
With skill73%
Checks it passed that the model alone missed
Includes response mapping in the flow verification
Configures explicit timeout limits for external HTTP requests
Implements retry mechanisms with idempotency keys for resilient external processing
and 5 more in the full report
5 fixed11 passed either way3 still failing3 broke
Pass-rate lift
+9 pp
Tokens
+37%
Agent turns
0%
Response time
+28%
Model gemini-3.6-flashMeasured 9/1/2026v116 of 22 cases passed with skill · 17 comparable
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 5 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 5
Prompt
Design an integration handler that calls an external REST endpoint for credit card charge processing. The team wants to execute a raw synchronous HTTP POST fetch without timeout limits or retry policies.
What was checked, without and with the skill
Checked
Without skill
With skill
Configures explicit timeout limits for external HTTP requests
Missed
Passed
Implements retry mechanisms with idempotency keys for resilient external processing
Missed
Passed
case-08 · gemini-3.6-flash · 9/1/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md27 lines2 files
Build Backend
Domain contract
Trace changed behavior from input through validation, authorization, domain
logic, persistence, side effects, and response mapping.
Preserve published contracts unless the request explicitly changes them.
Enforce authorization at a trustworthy server-side boundary.
Validate untrusted input before domain or persistence operations.
Keep multi-write operations atomic where partial state would violate invariants.
Design external calls for timeouts, retries, idempotency, and duplicate delivery where applicable.
Do not query or mutate production data without explicit authorization.
Do not fabricate migration, framework, or API behavior; verify version-sensitive details.
Avoid broad service abstractions that are used once.
Execute and verify
Read references/contracts-and-data.md when the task changes a public API, schema, transaction, job, or integration.
Implement the smallest change at the owning layer. Cover success, invalid
input, authorization, and material failure paths. Prefer:
reproducing/focused test → contract/integration test → type/lint → build
For migrations, validate forward behavior and document rollback or roll-forward strategy.