Completion reports can overlook failing checks or untested changes. This skill guides ordered execution of existing project scripts and reports failures, missing checks, and unverified scope.
Review the Skillstore skill "project-min-evaluation" from https://skillstore.io/skills/c3sc0-v4113-project-min-evaluation.md and its manifest at https://skillstore.io/api/skills/c3sc0-v4113-project-min-evaluation/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.
All 14 static findings are false positives caused by Markdown code spans, not Ruby or shell backtick execution. The workflow intentionally executes existing repository quality scripts, which require trusted project code. No evidence found of prompt injection, credential theft, or data exfiltration in the reviewed files.
c3sc0-v4113. (2026). project-min-evaluation security audit report (audit version 1) [Author version 1.0]. Skillstore. https://skillstore.io/skills/c3sc0-v4113-project-min-evaluation/audits/1
BibTeX 引用
@techreport{c3sc0-v4113-c3sc0-v4113-project-min-evaluation-2026,
author = {c3sc0-v4113},
title = {project-min-evaluation security audit report (audit version 1)},
institution = {Skillstore},
year = {2026},
number = {1},
url = {https://skillstore.io/skills/c3sc0-v4113-project-min-evaluation/audits/1},
note = {Author version 1.0}
}
CITATION.cff
cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "project-min-evaluation security audit report (audit version 1)"
version: "1.0"
type: report
authors:
- name: "c3sc0-v4113"
date-released: "2026-10-02"
url: "https://skillstore.io/skills/c3sc0-v4113-project-min-evaluation/audits/1"
identifiers:
- type: other
value: "skillstore:c3sc0-v4113-project-min-evaluation:audit:1"
description: "Skillstore immutable audit report identifier"
Skillstore 评分
评分依据证据置信度: 中
41
架构
85
可维护性
87
内容
65
社区
83
规范符合性
你能构建什么
Verify an implementation before handoff
Run available quality checks and identify any unverified scope before reporting a feature as complete.
Check a workspace change before commit
Run root checks and relevant package checks for touched workspace packages before committing.
Review verification evidence
Request exact failing commands, errors, and unavailable checks to assess whether a change is ready for review.
试试这些提示
Run basic quality checks
Use project-min-evaluation to run this repository's existing minimum quality checks. Report passed, failed, missing, and unavailable checks.
Verify before committing
Verify my changes before commit. Detect the package manager, run defined checks in order, and stop at the first failure.
Investigate a failed check
Run project-min-evaluation, diagnose the first failure, fix it within the change scope, and resume from that check. Report remaining unverified scope.
Evaluate a workspace behavior change
Evaluate this workspace change using root checks and checks for touched packages. Include test:e2e if behavior changed. Report lockfile conflicts and unavailable services.
最佳实践
Review repository script definitions and run them only in a trusted environment.
Resolve package manager conflicts before running checks.
Keep exact commands, errors, and unverified scope in the final verification report.
避免
Inventing commands for scripts that the project does not define.
Skipping a failing check to run later checks.
Reporting completion when required checks failed or could not run.
常见问题
Which assistants can use this skill?
The listing supports Claude, Codex, and Claude Code. Running checks requires repository access and command execution.
How does it choose the package manager?
The packageManager field takes priority. Otherwise, it uses pnpm, Bun, or npm lockfile signals, with npm as the fallback.
What happens when multiple lockfiles exist?
Without packageManager metadata, it reports the conflict and asks which manager the project uses.
Does it create missing quality scripts?
No. It reports missing scripts instead of inventing commands or adding configuration.
When does it run end-to-end tests?
It runs a defined test:e2e script only when end-to-end behavior changed.
What happens when a check fails or cannot run?
It stops progress through checks and reports the exact command, error, and unverified scope. Failed checks are fixed and rerun.