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/hoanganhduc/coding-system-rebuild --skill axiom-axle-mcp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。
ALWAYS use this skill when the user asks to send, get, retrieve, find, share, add, or search for a paper. This skill manages the user's Zotero library with 10,000+ papers and can retrieve PDFs, create share links, add new papers, and search. Prefer this over getscipapers for any request involving sending/getting/finding papers.
Offline runtime helper for loop ledgers plus headless drive, host-owned panel phases (--panel on, auto, or off), and the default cross-platform force-loop kit (bootstrap/start/drain with enforce/hard/notify defaults).
Run bounded autonomous research iterations with evidence gates, recovery ledgers, and optional cross-agent handoffs; prefers host-owned multi-agent panel with single-path drive primary; scripted force-loop defaults (Goal Focus enforce, hard goal_priority, notify ON).
基于 SOC 职业分类
正在显示 SKILL.md
| name | axiom-axle-mcp |
| description | Use when preparing optional AxiomMath AXLE MCP setup for manual formal-proof assistance. |
This skill is installed as an OpenCode-native SKILL.md. For runtime-backed
helpers, prefer the shared ai-agents-skills runtime root and the
AAS_RUNTIME_ROOT override instead of assuming a Codex-specific runtime
path.
On native Windows, use the managed Windows runner and the native runtime command target. Set $runtime to the installed runtime root. Multi-agent installs usually use %LOCALAPPDATA%\ai-agents-skills\runtime. Then run:
$runtime = if ($env:AAS_RUNTIME_ROOT) { $env:AAS_RUNTIME_ROOT } else { "$env:LOCALAPPDATA\ai-agents-skills\runtime" }
& "$runtime\run_skill.bat" "skills/axiom-axle-mcp/run_axiom_axle_mcp.bat" doctor
PowerShell runner target:
& "$runtime\run_skill.ps1" "skills/axiom-axle-mcp/run_axiom_axle_mcp.ps1" doctor
POSIX examples below use run_skill.sh and .sh command targets; use the Windows command target above on native Windows.
Use this skill only for explicit optional AXLE MCP setup. It never installs packages, starts an MCP server, writes MCP/client config, stores credentials, or calls AxiomMath services. It reports local readiness and emits manual configuration snippets with placeholders.
Check local readiness without running uvx:
bash "$AAS_RUNTIME_ROOT/run_skill.sh" \
skills/axiom-axle-mcp/run_axiom_axle_mcp.sh doctor
Emit a manual MCP config snippet:
bash "$AAS_RUNTIME_ROOT/run_skill.sh" \
skills/axiom-axle-mcp/run_axiom_axle_mcp.sh config-snippet
Run offline smoke:
bash "$AAS_RUNTIME_ROOT/run_skill.sh" \
skills/axiom-axle-mcp/run_axiom_axle_mcp.sh smoke
The emitted local stdio snippet uses command uvx and args ["--from", "axiom-axle-mcp==0.3.3", "axle-mcp-server"] with placeholder <AXLE_API_KEY>. The hosted URL https://mcp.axiommath.ai/mcp is manual setup only.
AXLE output is remote supplemental evidence. Record it as axle_remote_check, never as formal_check. It cannot set local lean_check_status, satisfy placeholder or trust-base scans, replace statement-equivalence review, or promote formal support without local Lean/project evidence.