Ready to supercharge your AI agents?

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

Verify Changes · agentscope-ai | DecimalAI

testingby agentscope-ai

Verify Changes

agentscope-ai/verify-changes

Source agentscope-ai/agentscope-java ↗License Not recorded

Verify code changes by running the project's typecheck, build, lint, and targeted tests, then fix and re-run until clean. Use after editing any source file.

NewMaintained

Automatic

Coding pack

Last verified 9/28/2026

What it changes

Passed 10 of 16 test cases with the skill, and 10 of 16 without it.

Without skill65%

With skill65%

Checks it passed that the model alone missed

Uses the -q flag for quiet output.
Suppresses terminal color output using NO_COLOR=1.
Refuses to mark work done without executing basic syntax validation.

and 2 more in the full report

4 fixed11 passed either way4 still failing4 broke

Pass-rate lift: 0 pp
Tokens: -26%
Agent turns: 0%
Response time: -54%

Model gemini-3.6-flashMeasured 9/28/2026v115 of 23 cases passed with skill · 16 comparable

Tokens, turns and time describe resource use, not a dollar price.

Use this skill

Save it into your project's skills folder. Claude Code picks it up in its next session.

curl -fsSL --create-dirs -o .claude/skills/agentscope-ai-verify-changes/SKILL.md https://app.decimal.ai/s/agentscope-ai-verify-changes/SKILL.md

Want it pinned and updatable? pip install "decimalai>=0.15", then decimalai skills add agentscope-ai-verify-changes, and start Claude Code with claude --plugin-dir "$(git rev-parse --show-toplevel)/.agents"

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

pip install "decimalai>=0.15"

2 · Get the skill — writes .agents/skills/agentscope-ai-verify-changes/ and pins it in .decimal/skills.lock

decimalai skills add agentscope-ai-verify-changes

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

curl -fsSL --create-dirs -o .claude/skills/agentscope-ai-verify-changes/SKILL.md https://app.decimal.ai/s/agentscope-ai-verify-changes/SKILL.md

Raw file: /s/agentscope-ai-verify-changes/SKILL.md ↗ · machine-readable summary at /s/agentscope-ai-verify-changes.json ↗

Safety checks: 3 of 3 passed

Code scan CleanIntent PassedContent Clean

See it in action

One of 4 test cases the skill fixed. Same model, same prompt, with and without the skill.

1 of 4

Prompt

I modified `app/models/user.py` and `tests/test_user.py`. The team usually runs tests on the entire repo. Which command should we run to check just these files cleanly without terminal colors?

What was checked, without and with the skill
Checked	Without skill	With skill
Specifies verification commands targeting app/models/user.py plus tests/test_user.py specifically.	Missed	Passed
Suppresses terminal color output using NO_COLOR=1.	Missed	Passed

case-08 · gemini-3.6-flash · 9/28/2026 · answers are capped and redacted

What's inside

The instructions your model reads when the skill loads.

SKILL.md23 linessingle file

Verify Changes

Editing a file is only half a step. Always close the loop by verifying before you mark work done or open a PR. This is the manual equivalent of an IDE/LSP diagnostics loop.

Procedure

Detect the project's check commands from its config files:
- Node/TS: package.json scripts (build, lint, typecheck), tsconfig.json -> npx tsc --noEmit, npx eslint <changed files>.
- Python: ruff check <path>, mypy <path>, pytest <specific test file>.
- Java/Maven: mvn -q -pl <module> compile, then the single relevant test class via -Dtest=<ClassName>.
- Gradle: ./gradlew :<module>:compileJava / :<module>:test --tests <ClassName>.
Run ONLY the checks and tests directly related to the files you changed. Never run the full test suite.
Disable color output so logs are readable: NO_COLOR=1, --no-color, or --no-colors.
Read the output as ground truth. If it reports errors, they are real — do not claim success.
Fix the root cause, then re-run the SAME command. Repeat until it passes.

Rules

Only mark a todo completed once its verification command passes.
If you cannot find a check command, at minimum compile/parse the changed files.
Do not disable or delete failing tests to make checks pass.

Track record

Benchmark

10 / 16

test cases passed, vs 10 / 16 without

Adoption

Users: 0
Times used: 0

Safety and source details

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v5 — 8/31/2026

View 15 scanner checks

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

agentscope-ai / agentscope-java

agentscope-examples/agents/agentscope-codingagent/src/main/resources/workspace-templates/skills/verify-changes/SKILL.md

View on GitHub ↗

Imported 8/31/2026Version v1

Other skills for the same job

Published skills whose descriptions closely match this one, with their headline benchmark lift.

More skills from the same source and nearby in the registry, with their headline benchmark lift.

Ready to supercharge your AI agents?

See it in action

What's inside

Verify Changes

Procedure

Rules

Track record

Benchmark

Adoption

SkillSafety

Source

Other skills for the same job

Related skills