Run project verification commands and report exit codes after code changes.
npx skills add https://github.com/t0lab/harness-kit --skill quality-gates-t0lab
Please help me install this Agent Skill. Skill: quality-gates Source: https://github.com/t0lab/harness-kit/tree/main/.agents/skills/quality-gates Command: npx skills add https://github.com/t0lab/harness-kit --skill quality-gates-t0lab
This Skill prevents premature completion claims by requiring fresh, executable evidence that changes behave as intended rather than relying on optimism, stale logs, or superficial inspection.
Run the project's appropriate verification commands for the changed files and report the fresh pass or fail evidence before declaring the work complete.
High-intent search queries and answers about installing and using this skill.
To enforce verification before marking a pull request done, you mandate running the project's appropriate tests, lint, or typecheck commands after the last code edit and reading their actual output as evidence.
A quality gate for code changes is a rule requiring fresh execution of verification commands like tests, builds, or lint, ensuring that completion claims are backed by actual passing exit codes rather than stale logs.
Choosing the right verification commands involves selecting the minimal honest set based on change type: tests for bug fixes, build and typecheck for API changes, and lint for style modifications.
Automated code review can require fresh test execution by enforcing a rule that mandates running verification commands covering touched files and reporting passing results or residual risk before declaring work complete.
After running typecheck and lint commands for a PR handoff, you should report the fresh passing or failing evidence, including exit codes, failure counts, and any remaining risk to prove the changes behave as intended.
Claiming a task is done without fresh verification causes integration issues because it relies on optimism or stale logs, bypassing the executable proof from tests or builds that changes actually behave as intended.
LinnkidChenAnalyze changed files to identify unit-test coverage gaps using project-specific specs.
jmdueaRun verification commands and check exit codes before declaring work complete.
growingioRun automated test commands and analyze output logs to validate code changes.
corticalstackValidate implementation plans against code changes and verification outcomes.
