observability-analyzer

Query and analyze Claude Code observability data (metrics, logs, traces). Use when analyzing performance, costs, errors, tool usage, sessions, conversations, or subagents.

Quality

76%

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

Fix and improve this skill with Tessl

tessl review fix ./skills/observability-analyzer/SKILL.md

Quality

Content

65%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 highly actionable query catalog with copy-paste-ready LogQL/PromQL examples, but it is held back by redundant query duplication, the absence of any sequenced workflow with validation checkpoints, and a monolithic inline reference structure that should be split into reference files.

Suggestions

Remove or collapse the 'Key Queries' section, which duplicates queries already shown under each analyze-* operation, to recover token budget.

Move the 'Event Types Reference' table and 'Key Queries' into a references/ file (e.g. references/event-types.md) and link to it from SKILL.md, keeping only a concise overview inline.

Add a short end-to-end workflow for generate-report (query → aggregate → validate counts → write report) with an explicit validation checkpoint before producing the final report.

Dimension	Reasoning	Score
Conciseness	The prose does not over-explain concepts Claude knows, but the "Key Queries" section substantially duplicates queries already shown in the Operations section (session_end, tool errors, context, subagent, skill, prompt), so it could be meaningfully tightened; not a 4 because the redundancy is non-trivial.	3 / 5
Actionability	Each operation ships a copy-paste-ready LogQL/PromQL query plus a concrete "Output" description, and the common analytic cases (errors, costs, tools, sessions, context) are all covered with executable examples.	5 / 5
Workflow Clarity	The body is a menu of independent queries rather than a sequenced multi-step process, and no validation or error-recovery checkpoints are given; each single operation is unambiguous, but there is no overarching workflow, matching the anchor with implicit/missing checkpoints.	3 / 5
Progressive Disclosure	Section structure and real script references (scripts/query-prometheus.sh, query-loki.sh, analyze-errors.sh, generate-report.sh — all present) exist, but the bulk reference material (Event Types table, Key Queries, Grafana dashboard list) is inlined in SKILL.md with no references/ split, so content that should be separate remains inline.	3 / 5
	Total	14 / 20 Passed

Description

87%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, third-person description that pairs a clear capability statement with an explicit, keyword-rich 'Use when' trigger list. Slightly limited verb variety keeps specificity and trigger quality at 4 rather than 5.

Dimension	Reasoning	Score
Specificity	"Query and analyze" names the domain plus enumerates several concrete analytic targets ("performance, costs, errors, tool usage, sessions, conversations, or subagents"), giving several specific capabilities with only minor verb-variety gaps; not a 5 because the action verbs themselves are limited to query/analyze.	4 / 5
Completeness	It explicitly answers both "what" ("Query and analyze Claude Code observability data (metrics, logs, traces)") and "when" ("Use when analyzing performance, costs, errors, tool usage, sessions, conversations, or subagents.") with concrete trigger phrases, matching the anchor exactly.	5 / 5
Trigger Term Quality	"performance, costs, errors, tool usage, sessions, conversations, or subagents" alongside "metrics, logs, traces" are natural user terms with synonyms, but a few natural variants (e.g. "telemetry", "logs") are less explicitly surfaced, keeping it just below comprehensive.	4 / 5
Distinctiveness Conflict Risk	"Claude Code observability data" carves a clear niche with distinct, telemetry-specific triggers and minimal overlap with other skills.	5 / 5
	Total	18 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
referenced_paths_exist	Referenced path issues: 1 missing	Warning

	Total	15 / 16 Passed

Repository: fernandezbaptiste/Skrillz
Path: skills/observability-analyzer/SKILL.md
Commit: 93ed392

Reviewed: 1 day 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.