gateway-setup
Set up a persistent AI agent gateway on macOS with Redis event bridge, heartbeat monitoring, and multi-session routing. Interactive Q&A to match your intent — from minimal (Redis + extension) to full (embedded daemon + Telegram + watchdog). Use when: 'set up a gateway', 'I want my agent always on', 'event bridge', 'heartbeat monitoring', 'agent notifications', or any request to make an AI agent persistent and reachable.
91
88%
Does it follow best practices?
Impact
96%
2.08xAverage score across 3 eval scenarios
Advisory
Suggest reviewing before use
Quality
Discovery
100%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is a strong skill description that excels across all dimensions. It provides specific concrete capabilities, includes an explicit 'Use when' clause with natural trigger terms, and occupies a very distinct niche that minimizes conflict risk. The description is well-structured and informative without being unnecessarily verbose.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: 'persistent AI agent gateway on macOS', 'Redis event bridge', 'heartbeat monitoring', 'multi-session routing', 'embedded daemon', 'Telegram', 'watchdog'. Also describes the interactive Q&A approach and the range from minimal to full setup. | 3 / 3 |
Completeness | Clearly answers both 'what' (set up a persistent AI agent gateway with Redis, heartbeat, multi-session routing, interactive Q&A for configuration) and 'when' (explicit 'Use when:' clause with multiple trigger phrases and a catch-all condition). | 3 / 3 |
Trigger Term Quality | Includes a strong set of natural trigger terms: 'set up a gateway', 'agent always on', 'event bridge', 'heartbeat monitoring', 'agent notifications', 'persistent and reachable'. These cover both specific technical terms and natural user phrasings. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive niche — persistent AI agent gateway on macOS with Redis event bridge is very specific and unlikely to conflict with other skills. The combination of macOS, Redis, heartbeat monitoring, and Telegram integration creates a clear, unique identity. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
77%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a highly actionable and well-structured skill with excellent executable code examples and clear multi-step workflows with validation checkpoints. Its main weakness is length — the monolithic structure packs ~400 lines into a single file where the tiered content could be split into separate referenced documents. The intent alignment Q&A section and decision chain table add useful context but contribute to token bloat.
Suggestions
Split Tiers 2-4 into separate referenced files (e.g., TIER2-HEARTBEAT.md, TIER3-MULTISESSION.md) and keep SKILL.md as a concise overview with Tier 1 inline and links to the rest.
Move the Decision Chain / ADR table to a separate ARCHITECTURE.md file — it's valuable context but not needed for setup execution.
Trim the Intent Alignment section to a compact decision matrix rather than full Q&A format, or move it to a separate PLANNING.md file.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is quite long (~400 lines) and includes some unnecessary context (e.g., explaining what each tier 'adds', the Credits section, the full ADR decision chain table). However, most content is concrete code/config rather than fluff. The intent alignment Q&A section adds bulk that could be trimmed. The Known Limitations section is valuable but verbose. | 2 / 3 |
Actionability | Excellent actionability throughout — fully executable TypeScript code for the extension, complete package.json, working bash commands for cron/Redis/tmux/launchd, and concrete verification commands. Code is copy-paste ready with real imports and proper error handling. | 3 / 3 |
Workflow Clarity | Each tier has clearly numbered build steps that build on the previous tier. The verification checklist at the end provides explicit validation checkpoints per tier. Critical setup notes at the top call out common failure modes (GATEWAY_ROLE, serveHost, ioredis issues) with specific fixes. The Known Limitations section honestly documents race conditions and recovery gaps. | 3 / 3 |
Progressive Disclosure | The tiered architecture (Tier 1-4) provides good conceptual progressive disclosure, and the ADR table references external files. However, all tiers are inline in a single massive file rather than split into separate documents. Tier 4 appropriately defers to an external article, but Tiers 1-3 could benefit from being separate files referenced from a concise overview. No bundle files are provided to support splitting. | 2 / 3 |
Total | 10 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 10 / 11 Passed | |
- Repository
- joelhooks/joelclaw
- Commit
03f0a59
- Reviewed
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.