Ready to supercharge your AI agents?

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

Get Started Free →

Project Verification · qierkang | DecimalAI

testingby qierkang

Project Verification

qierkang/project-verification

Source qierkang/platform-project-skill ↗License MIT

Verify changes or release readiness for this platform repository.

NewMaintained

Automatic

QA & Test Engineering pack

Last verified 8/28/2026

What it changes

Passed 10 of 16 test cases with the skill, and 12 of 16 without it.

Without skill75%

With skill63%

Checks it passed that the model alone missed

The response requires testing both default and standalone-infra Docker Compose configurations.
The response reads root and affected package.json files before reading references/verification-matrix.md.

2 fixed13 passed either way4 still failing5 broke

Pass-rate lift: -12 pp
Tokens: -34%
Agent turns: 0%
Response time: -48%

Model gemini-3.6-flashMeasured 8/28/2026v1 · latest is v215 of 24 cases passed with skill · 16 comparable

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

This benchmark measured an earlier version of the skill.

A newer attempt on gemini-3.6-flash on 10/4/2026 did not produce a publishable comparison (not comparable). The snapshot above is the previously measured run.

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/qierkang-project-verification/SKILL.md https://app.decimal.ai/s/qierkang-project-verification/SKILL.md

Want it pinned and updatable? pip install "decimalai>=0.15", then decimalai skills add qierkang-project-verification, 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/qierkang-project-verification/ and pins it in .decimal/skills.lock

decimalai skills add qierkang-project-verification

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

curl -fsSL --create-dirs -o .claude/skills/qierkang-project-verification/SKILL.md https://app.decimal.ai/s/qierkang-project-verification/SKILL.md

Raw file: /s/qierkang-project-verification/SKILL.md ↗ · machine-readable summary at /s/qierkang-project-verification.json ↗

Safety checks: 3 of 3 passed

Code scan CleanIntent PassedContent Clean

See it in action

One of 2 test cases the skill fixed. Same model, same prompt, with and without the skill.

1 of 2

Prompt

When starting verification for a modified backend package in our workspace, what is the initial sequence of inspection steps before choosing checks?

What was checked, without and with the skill
Checked	Without skill	With skill
The response reads root and affected package.json files before reading references/verification-matrix.md.	Missed	Passed

case-17 · gemini-3.6-flash · 8/28/2026 · answers are capped and redacted

What's inside

The instructions your model reads when the skill loads.

SKILL.md27 lines2 files

Project Verification

Verification must match the changed surface and produce real evidence.

Read Order

Read root and affected package build/test configuration (package.json in the default Node template); never assume scripts exist or impose the template stack on an existing project.
Read references/verification-matrix.md.
Inspect current changes to choose the narrowest checks that cover them.

Workflow

Run focused tests when the repository exposes an applicable test command.
Run the affected package build.
Expand to the root build for shared or cross-package changes.
Validate both default and standalone-infra Docker Compose configurations when deployment files change.
Use HTTP, browser, or screenshot evidence when the request changes runtime or visual behavior.
Report commands, results, and anything skipped or blocked.
If a task scope snapshot exists, run its check before declaring completion. A passing build cannot waive scope violations.
For database changes, follow the backend Skill's references/database-migrations.md: prove clean initialization, existing-data upgrade and repeated execution using the project's existing migration tool. Never report schema synchronization from generated files alone.

Guardrails

A passing typecheck is not runtime proof.
A healthy login or process is not proof that the requested workflow works.
Do not hide failed checks behind unrelated passing checks.
Do not claim coverage percentages unless a real coverage tool produced them.

Track record

Benchmark

10 / 16

test cases passed, vs 12 / 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 — 10/3/2026

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

qierkang / platform-project-skill

assets/templates/omni-platform/.agents/skills/project-verification/SKILL.md

View on GitHub ↗

License MITImported 10/3/2026Version v2Bundled 1 file

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

Project Verification

Read Order

Workflow

Guardrails

Track record

Benchmark

Adoption

SkillSafety

Source

Related skills