菜单
Manage and maintain AI coding skills via the skill-loop MCP server. Scan, inspect, review, fix, and monitor skill health.
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。
基于 SOC 职业分类
Synchronize ABI Markdown and docs claims with executable behavior. Use before editing README, AGENTS.md, CLAUDE.md, GEMINI.md, docs/**/*.mdx, walkthroughs, changelogs, or any old ABI Markdown that mentions WDBX, MCP, GPU, clustering, compression, FHE, agents, or benchmarks.
Validate ABI docs and external collateral against repo evidence and docs/contracts/external-claims-audit.mdx. Use when auditing claims, scanning docs for unsupported QPS/sharding/FHE/AES language, checking one claim before publish, or re-confirming the external-claims audit. Agent procedure only — not a CLI or slash command. Pair with abi-doc-claims-sync when editing docs.
Plan and execute ABI repo work from the TODO board, roadmap/spec docs, Zig 0.17 constraints, and validation gates. Use when the user asks to find ABI todos/roadmaps, organize a large implementation from tasks/todo.md, or decide the next safe work slice. Distinct from the goals skill — that skill owns the persistent ledger tasks/goals.md; this skill plans from the active board and north-star specs.
TUI/dashboard superpower. Launch interactive agent REPL, diagnostics dashboard panes, and slash commands via real abi CLI paths. Use when working on abi tui / dashboard / agent tui. There is no /abi-superpower-tui binary.
Build the abi CLI and non-interactively smoke the diagnostics dashboard — render `abi dashboard` one-shot (stdin from /dev/null forces the non-interactive fallback), assert exit 0 and all 5 panels (System, Plugins, WDBX Storage, Scheduler, Memory). Use to smoke-test `abi dashboard`/`abi tui`/`abi --tui` in CI or headless without hanging in the interactive TUI. Complements run-tui (which drives the interactive path via tmux).
Capture, track, and execute goals for the abi Zig project. Use when the user wants to record a goal, see outstanding goals, update goal status, or break a goal into a verified work slice. Distinct from abi-goal-orchestrator (which plans from the TODO board) — this skill owns a persistent goal ledger and drives goal-to-slice execution.
| name | sl |
| description | Manage and maintain AI coding skills via the skill-loop MCP server. Scan, inspect, review, fix, and monitor skill health. |
Route the user's request to the skill_loop MCP tool. This skill is a convenience wrapper -- all actions are handled by the MCP server.
/sl <action> where action is one of:
| Action | What it does |
|---|---|
scan | Scan for SKILL.md files and register them |
status | Health dashboard: skill count, runs, failure rate |
review | Analyze all skills for failure patterns and staleness |
fix | Propose amendments for broken skills (creates a git branch) |
fix --dry-run | Preview fixes without modifying anything |
list | Show all registered skills with metadata |
runs | Show recent skill run activity |
history | List past amendments and their status |
detection | Show detection stats and active sessions |
gc | Prune old run data |
/sl)skill_loop MCP tool with action set to the parsed actionIf no action is provided, default to status.
If the MCP server is not connected, tell the user to add this to their .mcp.json:
{
"mcpServers": {
"skill-loop": {
"command": "npx",
"args": ["-y", "-p", "@stylusnexus/skill-loop-cli", "skill-loop-mcp"]
}
}
}
Note: npx-based MCPs may fail handshake in some envs (see startup diagnostics). Alternatives or local installs can be configured in .mcp.json. After editing MCP config, restart the Grok session/TUI.
CLI vs MCP: There is no in-repo skill-loop binary. The CLI ships as
@stylusnexus/skill-loop-cli (repo pins @0.3.3 in .mcp.json). Bare
skill-loop on PATH is optional — use
npx -y -p @stylusnexus/skill-loop-cli@0.3.3 skill-loop <cmd> or
npm i -g @stylusnexus/skill-loop-cli@0.3.3. CLI subcommands differ from the
MCP /sl actions above: rescan is init (not scan); analysis is inspect;
logging is skill-loop log <skill-name> <outcome>.
During 2026-07-06 "do all" / hygiene pass, used skill-loop MCP directly for review + amend + apply_fixes on registered skills (fixed broken file refs in sl and abbey-assistant). Full-check now passes (45/45). Added cross-refs in other skills. Re-scan after adding new SKILL.md. See ~/CLEANUP-2026-07-06.md.