MCP App StoreMCP App Store
from antigravity-skills868
Design and operate remote sandboxed agent runtimes with warm pools, snapshot/restore, per-session isolation, and user-authored PR workflows for scalable backgro
Hosted Agent Infrastructure codifies best practices for running agents in remote, sandboxed environments. It focuses on eliminating cold-start latency via image registries and warm pools, snapshot/restore for fast follow-ups, per-session state isolation, and an API-driven flow to safely extract outputs (branches, PRs, logs) using user tokens. The skill explains architecture layers — sandbox, API, and client — and concrete operational strategies like predictive warm-up, image rebuild cadence, and git identity management for background commits.
Activate this skill when building background coding agents, multi-client collaborative sessions, or systems that must scale agent execution beyond a single machine. Use it for designs where session speed, reproducibility, and auditability (user-authored PRs) are primary concerns. Avoid for unrelated topics like evaluator governance or tool contract design.
scripts/ (build/warm-pool and snapshot patterns)references/Reasonably compatible with server-first agent platforms and CLIs (Claude Code, Cursor, Copilot-style integrations) that can run external scripts and manage long-lived sessions.
Hosted Agent Infrastructure skill provides architectural guidance for building sandboxed agent runtimes with warm pools, snapshot/restore, and per-session isolation. The bundled sandbox_manager.py is well-structured pseudocode with clean class separation but all core methods are stubs (pass). The demo crashes because _cold_start returns None and start_session tries to use it. SKILL.md is thorough and well-organized with clear triggers, gotchas, and integration notes.
Well-written architectural skill with good SKILL.md structure. The script is reference pseudocode, not production code. Shell injection in git config commands is the main security concern but unlikely to be exploitable in the skill's intended context (sandbox environments). No malicious intent detected.
NGINX Integration
Integrate with NGINX via the Membrane CLI to query, manage, and proxy requests to NGINX for configuration, logs, and runtime operations.
Starlark Dev
Create, debug, and test Kurtosis Starlark packages — write package structure, run dry-runs, and inspect plan execution for reliable orchestration.
Sync Production Database to Development
Run a safe, scripted workflow to download production database dumps and restore them into a development environment (download-only and restore options).
Sandbox0 Integration
Guidance and templates for integrating Sandbox0 sandboxing into AI agents — CLI/SDK patterns, templates, volumes, network policy, and deployment choices.
Alpha Forge Pre-Ship Quality Gates
Pre-merge quality gates for PRs that validate RNG determinism, forked URLs, runtime parameter ranges, and manifest synchronization to reduce review cycles.
JJ Hunk
Programmatic hunk selection for Jujutsu (jj): list, split, commit, or squash specific hunks to enable partial commits and fine-grained history editing without a
Overnight — Autonomous Long-Running Coding
Orchestrates long-running coding goals: decomposes objectives into atomic tasks, dispatches isolated worktree workers, verifies acceptance criteria, and merges
Skill Issue — Skill Auditor
Scan and audit installed agent skills to produce a health report covering inventory, usage, dependency checks, version status, and actionable recommendations.