CtrlK
BlogDocsLog inGet started
Tessl Logo

gateway-diagnose

Diagnose gateway failures by reading daemon logs, session transcripts, Redis state, and OTEL telemetry. Full Telegram path triage: daemon process → Redis channel → command queue → pi session → model API → Telegram delivery. Use when: 'gateway broken', 'telegram not working', 'why is gateway down', 'gateway not responding', 'check gateway logs', 'what happened to gateway', 'gateway diagnose', 'gateway errors', 'review gateway logs', 'fallback activated', 'gateway stuck', or any request to understand why the gateway failed. Distinct from the gateway skill (operations) — this skill is diagnostic.

74

Quality

93%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

86%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.

A tight, project-specific diagnostic runbook that respects the token budget and encodes real operational judgment (ownership preservation, evidence discipline, secret hygiene). Its main gap is that the executable diagnostic surface is delegated to the CLI's --help output rather than surfaced in the skill itself.

Suggestions

Surface one or two of the most common diagnostic commands (beyond the --help pointer) inline, e.g. the status and event-inspection invocations used in the typical triage, so the first probe can be executed without an extra discovery step.

Make the implicit sequence explicit — renumber the paragraphs as a short ordered diagnostic checklist (identify → place → trace → isolate → verify/report) so the workflow order and the verify step stand out.

Add a brief feedback-loop line for the trace step (e.g. what to do when the first correlation id dead-ends: advance to the next hop in the path rather than stopping), matching the 'do not stop the whole task' guidance with a concrete recovery action.

DimensionReasoningScore

Conciseness

The ~15-line body is lean prose with zero padding: no concepts Claude already knows are explained, and every sentence adds project-specific guidance ("Queue age alone does not prove a hung agent", "A successful queue submission is not proof of delivery").

5 / 5

Actionability

One concrete command is given ("Use `joelclaw gateway --help` for supported status, event, and diagnostic commands") along with specific comparison artifacts ("timestamps, correlation ids, progress, and configured timeouts"), but the actual diagnostic commands are deferred to --help rather than enumerated, leaving minor gaps.

4 / 5

Workflow Clarity

A clear diagnostic sequence runs through the paragraphs (find the failing id → read entrypoint/config/runbook → resolve service placement → trace the path → follow the first relevant failure → "Verify the affected behavior and report the exact observation"), with a verification checkpoint present, but the sequence is implicit in prose rather than an explicit ordered list and there is no defined error-recovery loop.

4 / 5

Progressive Disclosure

The skill is well under 50 lines with no external references needed, and the short paragraphs are logically ordered under a single heading, so the simple-skill exception applies; nothing that belongs in a separate file is inlined.

5 / 5

Total

18

/

20

Passed

Description

100%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.

The description is a strong example: concrete capabilities, an explicit and thorough Use-when trigger list, and explicit boundary marking against a sibling operations skill. It is concise despite its detail, with no fluff or over-claims.

DimensionReasoningScore

Specificity

"Diagnose gateway failures by reading daemon logs, session transcripts, Redis state, and OTEL telemetry" plus the enumerated path "daemon process → Redis channel → command queue → pi session → model API → Telegram delivery" lists multiple specific concrete actions with comprehensive coverage of the diagnostic data sources and trace path.

5 / 5

Completeness

It explicitly answers both what (diagnose gateway failures by reading four named evidence sources, with a full path triage) and when ("Use when: 'gateway broken', ... or any request to understand why the gateway failed") with concrete trigger phrases, exactly matching the top anchor.

5 / 5

Trigger Term Quality

Eleven quoted natural trigger phrases ("gateway broken", "telegram not working", "why is gateway down", "check gateway logs", "fallback activated", "gateway stuck") plus a catch-all clause give comprehensive coverage of terms users would naturally say, including synonyms and phrasings of varying formality.

5 / 5

Distinctiveness Conflict Risk

"Distinct from the gateway skill (operations) — this skill is diagnostic" explicitly disambiguates against the most likely competing skill, and the diagnostic niche with its own trigger vocabulary carries minimal conflict risk.

5 / 5

Total

20

/

20

Passed

Validation

93%

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

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

15

/

16

Passed

Repository
joelhooks/joelclaw
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.