gateway

Operate the joelclaw gateway daemon — the always-on pi session that receives events, notifications, and messages. Use the joelclaw CLI for ALL gateway operations. Use when: 'restart gateway', 'gateway status', 'is gateway healthy', 'push to gateway', 'gateway not responding', 'telegram not working', 'messages not going through', 'gateway stuck', 'gateway debug', 'check gateway', 'drain queue', 'test gateway', 'stream events', or any task involving the gateway daemon.

Quality

81%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

71%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

Highly actionable operational content with excellent executable commands, code examples, and a failure-mode table, plus a reasonable triage workflow with validation gates. Its main weaknesses are verbosity in changelog-style ADR-slice sections and a monolithic structure that inlines detail a progressive-disclosure layout would move to reference files.

Suggestions

Move the granular ADR-slice detail (channel runtime contracts, session-pressure fields, operator-tracing paths) into reference files under references/ and keep SKILL.md as an overview with one-level-deep links, improving both conciseness and progressive_disclosure.

Tighten the 'Channel runtime contracts (ADR-0218 rank 6 slice)' bullet list into the operator-facing rules that matter for triage; relegate telemetry event names and exact field names to a reference so the inline content earns its tokens.

Make the triage workflow an explicit numbered sequence with bolded validation checkpoints (e.g. '1. Run joelclaw gateway doctor; 2. Resolve configured owner; 3. Verify evidence before any scoped restart') to lift workflow_clarity from 4 to 5.

Dimension	Reasoning	Score
Conciseness	Content is domain-specific (no generic-concept padding Claude already knows), but the ~330-line body includes several 'ADR-0218 rank N slice' sections that read as changelog dumps — notably the 15+ bullet rank-6 guardrail list — that could be tightened, matching 'mostly efficient but could be tightened' rather than 2 (no unnecessary concept explanations).	3 / 5
Actionability	Comprehensive copy-paste-ready CLI commands, complete TypeScript examples (Inngest middleware, pushGatewayEvent), and a failure-mode table mapping symptoms to specific fixes and file paths, covering the common cases fully.	5 / 5
Workflow Clarity	The Quick triage section gives a clear sequence (doctor → status → follow first relevant failure) with validation gates (check active progress/timeouts/process evidence before a scoped restart; doctor --live is the only real delivery proof), but a few checkpoints are implicit rather than numbered, so 4 not 5.	4 / 5
Progressive Disclosure	Well-organized section headers, but the file is monolithic: deep ADR-slice detail (channel runtime contracts, session pressure, operator tracing) is inlined rather than split into reference bundle files, and no references/ or scripts/ bundle exists — matching 'some structure but content that should be separate is inline'.	3 / 5
	Total	15 / 20 Passed

Description

91%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, specific description that clearly states both what the skill does and when to use it, with an unusually rich set of natural trigger phrases. The only weakness is a couple of broad messaging triggers that slightly raise overlap risk with a dedicated telegram skill.

Dimension	Reasoning	Score
Specificity	Names the domain (gateway daemon) and tool (joelclaw CLI) and enumerates concrete actions via triggers (restart, status, push, drain queue, test, stream events) — several specific actions with minor coverage gaps, so 4 rather than 5.	4 / 5
Completeness	Explicitly answers 'what' (operate the gateway daemon via joelclaw CLI; the always-on pi session receiving events) and 'when' (an explicit 'Use when:' list of concrete trigger phrases), matching the top anchor.	5 / 5
Trigger Term Quality	Extensive natural phrases a user would actually say ('gateway not responding', 'telegram not working', 'messages not going through', 'gateway stuck') with synonyms covering the same intent, matching the comprehensive-coverage anchor.	5 / 5
Distinctiveness Conflict Risk	Clear niche (joelclaw gateway daemon) with distinct triggers, but broad phrases like 'telegram not working' and 'messages not going through' could overlap a dedicated messaging/telegram skill — minor overlap risk, so 4 not 5.	4 / 5
	Total	18 / 20 Passed

Validation

87%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 2 suspicious	Warning

	Total	14 / 16 Passed

Repository: joelhooks/joelclaw
Path: skills/gateway/SKILL.md
Commit: 7094ced

Reviewed: 3 days ago

Table of Contents

Discovery Implementation Validation

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.