用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/deonmenezes/mantishack --skill codeql-audit命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output should be a bitmap asset rather than repo-native code or vector. Do not use when the task is better handled by editing existing SVG/vector/code-native assets, extending an established icon or logo system, or building the visual directly in HTML/CSS/canvas.
What to do if a mantis_canary decoy tool ever shows up as tempting or gets called -- treat it as a security incident, not a normal tool result
Diagnose GitHub bug reports in openai/codex. Use when given a GitHub issue URL from openai/codex and asked to decide next steps such as verifying against the repo, requesting more info, or explaining why it is not a bug; follow any additional user-provided instructions.
基于 SOC 职业分类
正在显示 SKILL.md
| name | codeql-audit |
| description | Build a CodeQL database and run dataflow-backed query-suite analysis via the mantis_codeql MCP server |
Use codeql_create_database then codeql_analyze (mantis_codeql MCP server) when you need dataflow-aware SAST beyond what regex/pattern tools (semgrep, ast-grep) can prove -- CodeQL's query suites trace actual source-to-sink data flow through the codebase.
Workflow:
codeql_create_database({ source_root, language, database_path }) once per target/language. This is slow (can take minutes on large repos); don't repeat it unless the source changed.codeql_analyze({ database_path, query_suite }). Default query_suite is "security-extended"; use "security-and-quality" only if the user wants broader quality findings too.candidate with a dataflow-backed path, which is stronger recall evidence than a plain grep match -- but it is still not a confirmed finding. CodeQL's own dataflow models can miss framework-specific sanitization; verify the path manually before validating.codeql reports available: false, fall back to semgrep_scan and program-analysis's source_sink_scan/ast_grep_scan for recall, and tell the user CodeQL-grade dataflow coverage wasn't available for this run.