菜单
Run Foreman repo-local agent simulation validation scenarios for PR-sized behavior checks, especially synthetic tmux/native-signal/Pi-subagent rollouts that should exercise public Foreman CLI surfaces without requiring real long-running agents.
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
Use when designing, reviewing, or changing Foreman's native Swift macOS overlay UI. Applies macOS HIG guidance, Foreman-specific command-palette UX, screenshot review, accessibility, and visual iteration rules.
Use when adding tests, fixtures, screenshot capture, fake foreman harnesses, or validation tasks for Foreman's Swift macOS overlay.
Use when designing or building native macOS applications with SwiftUI or AppKit. Triggers on menu bar structure, keyboard shortcuts, multi-window behavior, Liquid Glass design system, macOS Tahoe/Sequoia, sidebar navigation, toolbar design, app icons, SF Symbols, or making an app feel like a "good Mac citizen."
Apple Human Interface Guidelines for Mac. Use when building macOS apps with SwiftUI or AppKit, implementing menu bars, toolbars, window management, or keyboard shortcuts. Triggers on tasks involving Mac UI, desktop apps, or Mac Catalyst.
Creates, updates, or evaluates AGENTS.md and CLAUDE.md files for a repository. This skill activates when users want to create onboarding docs, update agent context files, or evaluate documentation quality. Generates docs with YAML frontmatter indexes for machine-readable progressive disclosure, system invariants, redirect-pattern gotchas (DO/NOT/BECAUSE), and validated commands. Supports incremental updates via git-based staleness detection.
Comprehensive documentation workflow for creating, generating, and maintaining technical documentation. This skill activates when users want to analyze documentation needs, generate module quickstart guides, sync docs with code changes, or consolidate scattered docs. Supports four modes - analyze (plan docs), quickstart (generate onboarding docs), update (sync with code changes), and consolidate (inventory and tidy existing docs).
| name | foreman-agent-sim-validation |
| description | Run Foreman repo-local agent simulation validation scenarios for PR-sized behavior checks, especially synthetic tmux/native-signal/Pi-subagent rollouts that should exercise public Foreman CLI surfaces without requiring real long-running agents. |
Use this skill when validating Foreman behavior that benefits from a realistic operator simulation but should remain deterministic, local, and CI-friendly. These scenarios are agent simulations, not dogfood studies: they use scripted fixtures and public Foreman commands. Reserve dogfood for real worker-prompt or subagent replay studies where agent behavior itself is under observation.
foreman agents, foreman extensions, and
focused integration tests over unit-only seams.agent_sim, smoke, or rollout; do not call them
dogfood.Current scenarios live under scenarios/:
pi-subagent-activity/ — synthetic Pi pane plus fresh pi-subagents status
files. Proves native Pi subagent activity promotion, extension cards, and
one-way workspace containment.Pick the scenario matching the changed behavior.
Read the scenario README.md and expected-observations.md.
Run setup.sh when present.
Run the scenario command or test exactly as documented.
Run collect.sh when present to gather compact observations.
Record validation with HK, for example:
hk validate --why "Agent simulation proves <behavior>." -- <scenario-command>
If adding a new scenario, include at least:
scenarios/<name>/
README.md
setup.sh
collect.sh
expected-observations.md
Add worker-prompt.md only when the scenario intentionally asks a real
worker/subagent to perform a replay. Otherwise keep it scripted.
Run the Pi subagent activity simulation:
cargo test --test pi_subagent_agent_sim
Run the focused unit seam it depends on:
cargo test -q pi_subagents