CategoryWeightCategory scoreContribution
Securityprompt, exec, net, exfil, eval
35%
22.4 pts
Supply chainhash, typosquat, maintainer, lockfile
20%
20.0 pts
Maintenancestaleness, pinning, CI
15%
15.0 pts
TransparencySKILL.md, perms, README
15%
15.0 pts
Communityinstalls, verify, response
15%
15.0 pts
Findings & checks · 3 flagged
Security
MEDIUMInstruction telling the agent not to ask for approvalSS-SKILL-INJECT-DONT-ASK-01 · Prompt injection · .cursor/commands/trellis-brainstorm.md×3
MEDIUM — it fires on intent; the real damage depends on the host agent's own approval-gating.
Why it matters
The text {match} tells the agent to skip the normal "ask the user first" gate. Used adversarially it removes the human-in-the-loop check before destructive or sensitive actions, turning a normally-gated agent into a fire-and-forget executor.
The exact value spotted
42
437. **No meta questions**
44Do not ask "should I search?" or "can you paste the code so I can continue?"
45If you need information: search/inspect. If blocked: ask the minimal blocking question.
46
Occurrences
3 occurrences · first at L44, also L157, L167Show all 3 locations
Line
File
L44
.cursor/commands/trellis-brainstorm.md
L157
.cursor/commands/trellis-brainstorm.md
L167
.cursor/commands/trellis-brainstorm.md
How to fix
Remove the approval-skipping instruction, or scope it narrowly to a specific safe, reversible action.
- Delete blanket "don't ask / no need to confirm" directives from the skill.
- If the skill is a genuine autonomous job, restrict the opt-out to a named non-destructive action rather than all actions.
Framework references
OWASPLLM01ATLASAML.T0051Trace & refs
Supply chain
✓—All supply chain checks passedNo findings in this category for the latest scan.pass
Maintenance
✓—All maintenance checks passedNo findings in this category for the latest scan.pass
Transparency
✓—All transparency checks passedNo findings in this category for the latest scan.pass
Community
✓—All community checks passedNo findings in this category for the latest scan.pass
Every scanned point with the score it earned and what moved between them.
d997b2c
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
qcqx-project-manage
一个更可靠、方便的项目管理工具，以树形结构组织项目，支持拖拽排序、关联链接、多窗口同步，还能自动扫描本地 Git 仓库。
项目结构
本项目为 pnpm monorepo，包含以下子包：
packages/
├── core/ # 共享核心库 — Git 扫描、配置管理、项目树数据结构
├── vscode/ # VS Code / Cursor 插件
├── cli/ # 命令行工具 (qpm)
└── mcp/ # MCP Server — 通过 Model Context Protocol 暴露项目管理能力
skills/ # AI Agent Skill — 纯文档，适配 Cursor / Claude Code 等平台| 包 | 说明 | 详情 |
|---|---|---|
@qcqx/project-manage-core | 共享核心逻辑：Git 项目扫描、状态检测、配置与缓存管理、项目树操作 | README |
qcqx-project-manage | VS Code 插件，提供项目列表、最近文件夹、本地 Git 项目三个视图 | README |
@qcqx/project-manage-cli | 命令行工具 qpm，在终端中管理项目 | — |
@qcqx/project-manage-mcp | MCP Server，让 AI agent 通过标准协议使用项目管理功能 | README |
| AI Agent Skill | 纯文档 Skill，适配 Cursor / Claude Code 等平台 | README |
开发
# 安装依赖
pnpm install
# 构建所有包
pnpm build
# 仅构建 core
pnpm build:core
# 仅构建 VS Code 插件
pnpm build:vscode
# 开发模式（watch）
pnpm watch
# 类型检查
pnpm check-types
# 打包 VS Code 插件为 .vsix
pnpm package:vsix
# 发布 VS Code 插件
pnpm publish:vscode环境要求
- Node.js ≥ 16
- pnpm ≥ 9
QA
为什么支持文件保存？
有很多配置类的文件分散在各处，比如 ssh、claude 的配置，保存起来，快速打开。