CtrlK
BlogDocsLog inGet started
Tessl Logo

error-analysis

Help the user systematically identify and categorize failure modes in an LLM pipeline by reading traces. Use when starting a new eval project, after significant pipeline changes (new features, model switches, prompt rewrites), when production metrics drop, or after incidents.

64

Quality

76%

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

Fix and improve this skill with Tessl

tessl review fix ./skills/error-analysis/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

75%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.

The content is a well-structured, actionable methodology with concrete templates, code, and prompts and a clear sequenced workflow. It is efficient without being padded, with only minor opportunities to tighten prose and add an explicit verification checkpoint.

DimensionReasoningScore

Conciseness

The body is largely lean, actionable methodology prose with minimal padding and no exposition of concepts Claude already knows; a few explanatory asides (e.g. "errors cascade, so downstream symptoms disappear...") could be trimmed, placing it just below the leanest 5 anchor.

4 / 5

Actionability

Provides concrete, copy-paste-ready artifacts (a notes table template, a Python failure-rate snippet, and an LLM-clustering prompt) plus a specific fix-decision tree, with only minor gaps keeping it just below fully comprehensive.

4 / 5

Workflow Clarity

A clear 7-step sequence (Collect → Read → Group → Label → Compute → Decide → Iterate) with a feedback loop and stopping criteria; this is analysis rather than destructive/batch work, and the only minor gap is the absence of an explicit verification checkpoint on the computed failure rates before prioritization.

4 / 5

Progressive Disclosure

No bundle files exist and the single SKILL.md is well-organized into clearly headed sections with inline cross-references to other skills rather than nested bundle references; it exceeds the simple-skill 50-line threshold, so it sits at 4 rather than 5.

4 / 5

Total

16

/

20

Passed

Description

78%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 clearly and explicitly covers both capability and trigger conditions with concrete, natural language. It is slightly light on enumerated specific actions, which keeps specificity at the midpoint rather than the top.

DimensionReasoningScore

Specificity

Names the domain (LLM pipeline failure modes) and two concrete actions ("identify and categorize" failure modes "by reading traces"), but does not enumerate several specific actions, matching the 1-2 concrete actions anchor rather than the comprehensive 4-5 anchors.

3 / 5

Completeness

Explicitly answers both what ("identify and categorize failure modes in an LLM pipeline by reading traces") and when ("Use when starting a new eval project, after significant pipeline changes...") with concrete trigger phrases, matching the top anchor.

5 / 5

Trigger Term Quality

Good natural trigger coverage ("starting a new eval project", "production metrics drop", "after incidents", "new features, model switches, prompt rewrites") that users would plausibly say, though a few common variations (e.g. "debugging", "evals") are absent.

4 / 5

Distinctiveness Conflict Risk

The LLM-pipeline-trace failure-analysis niche is mostly distinct with specific triggers, but could overlap slightly with general debugging or eval-building skills, so it sits at 4 rather than a fully conflict-free 5.

4 / 5

Total

16

/

20

Passed

Validation

100%

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

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
hamelsmu/evals-skills
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.