cost-anomaly-detection

Use when proactively scanning for cost anomalies, unusual spending, unexpected charges, or irregular patterns — during weekly reviews, after incidents, or when something looks off

1.25x

Quality

66%

Does it follow best practices?

Impact

100%

1.25x

Average score across 3 eval scenarios

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./plugins/cost-analyst/skills/cost-anomaly-detection/SKILL.md

Quality

Content

50%Weight 40%Scale 1-3

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 thorough, well-sequenced detection procedure with concrete API calls and statistics code, plus a useful 'all math in code' guardrail. Its weaknesses are length/padding (especially the placeholder output template), some pseudocode and placeholder data binding, and a monolithic structure that inlines material better suited to separate reference files.

Suggestions

Move the ~200-line output-format template into a separate reference file (e.g. references/output-format.md) and keep only a short summary plus a link in SKILL.md to improve conciseness and progressive disclosure.

Replace the Step 9 pseudocode and `[...]` data-binding placeholders with a complete executable Python snippet to raise actionability toward level 3.

Add an explicit validation checkpoint (e.g. 'Before reporting, cross-check each flagged anomaly against org context and mark false positives') as a workflow step to lift workflow clarity.

Dimension	Reasoning	Score
Conciseness	The body is mostly actionable domain guidance rather than beginner explanations, but at ~620 lines it carries significant padding — notably a ~200-line output template of placeholder text and catalogs (Common Anomaly Types, Tips) that restate knowledge Claude already has.	2 / 3
Actionability	Concrete `get_cost_data(...)` calls and executable statistics code (mean, stdev, z-score) are provided, but Step 9 is pseudocode and data binding is left as `[...]` placeholders, leaving key details incomplete per the level-2 anchor.	2 / 3
Workflow Clarity	An explicit 11-step sequence is clearly laid out, but there are no explicit validation checkpoints or fix-retry feedback loops within the workflow; false-positive validation lives implicitly in the output-format section rather than as a workflow gate.	2 / 3
Progressive Disclosure	Sections are well-organized and external references are clearly signaled one level deep in 'See Also', but the file is monolithic — the large output template, anomaly-type catalog, and advanced techniques are inlined in SKILL.md rather than split into reference files.	2 / 3
	Total	8 / 12 Passed

Description

82%Weight 40%Scale 1-3

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 trigger-rich and clearly answers both what the skill does and when to use it, with strong natural-language phrasing. Its main weakness is that it leads with the 'Use when' framing and names only one action ('scanning') rather than enumerating multiple concrete capabilities, leaving it somewhat generic relative to the best examples.

Suggestions

Lead with a third-person action statement before the trigger clause, e.g. 'Detects and classifies cost anomalies across services, accounts, regions, and resources. Use when...'.

Add 1-2 more concrete actions (e.g. 'classifies anomaly severity', 'flags potential security and waste indicators') to lift specificity toward level 3.

Tighten distinctiveness by contrasting with general cost analysis, e.g. 'Use for anomaly detection specifically, not general cost reporting or optimization.'

Dimension	Reasoning	Score
Specificity	Names the cost-anomaly domain and the action 'scanning' applied to several objects (anomalies, unusual spending, unexpected charges, irregular patterns), but lists a single action rather than multiple distinct concrete actions, so it does not reach level 3.	2 / 3
Completeness	Both 'what' (proactively scanning for cost anomalies) and 'when' (during weekly reviews, after incidents, or when something looks off) are explicit, and an explicit 'Use when' trigger clause is present, satisfying the level-3 requirement.	3 / 3
Trigger Term Quality	Strong coverage of natural phrases a user would actually say — 'cost anomalies', 'unusual spending', 'unexpected charges', and 'when something looks off' — matching the level-3 anchor of good natural-term coverage.	3 / 3
Distinctiveness Conflict Risk	The anomaly-detection niche is reasonably clear, but triggers like 'unusual spending' and 'unexpected charges' could still overlap with broader cost-analysis or cost-review skills, so it is not unambiguously distinct.	2 / 3
	Total	10 / 12 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
skill_md_line_count	SKILL.md is long (622 lines); consider splitting into references/ and linking	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: Cloudzero/cloudzero-claude-marketplace
Path: plugins/cost-analyst/skills/cost-anomaly-detection/SKILL.md
Commit: b9c7c16

Reviewed: 13 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.