القائمة
Monitor CI status, GitHub stats, and repo health for Aegis. Use for /loop CI watching, post-push verification, and launch metrics tracking.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
AEGIS project context — Electron desktop app for AI agent monitoring with Svelte 5 renderer. Provides architecture map, module locations, IPC channels, component inventory, and coding conventions. MUST be invoked before working on any AEGIS source file: editing scanners, watchers, or scoring in src/main/; modifying Svelte components, stores, or utils in src/renderer/; updating agent-database.json, constants.js, or types in src/shared/; adding IPC channels to preload.js; debugging anomaly-detector, rule-loader, network-monitor, or process-scanner; writing tests; or asking how AEGIS architecture works. Skip only for trivial one-line fixes (typos, renames) and bare git/npm commands.
AEGIS-specific pre-push / pre-release repo audit. Runs format:check, build:renderer, lint, agent-database count, README cross-check, git status, and tracked-internals check. Use when the user says "run audit", "check repo health", or "pre-push check" inside the AEGIS project.
AEGIS-specific post-task gate — runs format:check, build:renderer, lint, then stages only the files changed by the task (never `-A`), conventional commit, push to the current feature branch (never master). Use when the user says "commit and push", "ship it", or "track this commit" inside the AEGIS project.
Electron main process patterns for Aegis. CJS modules, platform abstraction, IPC, file watchers. Use when editing src/main/ files.
Vitest testing patterns for Aegis. ESM imports, mocking, test structure. Use when writing or editing tests.
Aegis prompt formula for Claude Code and Antigravity. Use when writing prompts, planning tasks, starting new work, or when the user says "write a prompt", "create a task", "plan work", or asks how to structure a request for an AI coding assistant. Also use proactively when decomposing a large request into atomic subtasks.
| name | ci-monitor |
| description | Monitor CI status, GitHub stats, and repo health for Aegis. Use for /loop CI watching, post-push verification, and launch metrics tracking. |
gh run list --repo antropos17/Aegis --limit 3
gh run view $RUN_ID
Wait for status != "in_progress", then report PASS/FAIL.
gh api repos/antropos17/Aegis → stars, forks, open issuesgh issue list --state open --limit 20 → open issuesgh pr list --state open → open PRsnpm audit --audit-level=high → security/loop 2m watch CI for latest push. Stop when all checks pass or fail.
/loop 1h track stars/forks/issues count on Aegis. Log changes.
/loop 1d run npm audit. Flag HIGH/CRITICAL.
5 open PRs without review → MEDIUM