agent-sort

Build an evidence-backed ECC install plan for a specific repo by sorting skills, commands, rules, hooks, and extras into DAILY vs LIBRARY buckets using parallel repo-aware review passes. Use when ECC should be trimmed to what a project actually needs instead of loading the full bundle.

Quality

81%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

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.

A well-structured, largely actionable instruction skill: concrete commands, formats, and a sequenced workflow with an explicit verification step. The main gaps are the absence of executable verification checks and an error-recovery loop, and a monolithic single-file layout that could lean out some peripheral detail.

Suggestions

Add concrete executable verification commands to step 6 (e.g. a loop that checks each DAILY path exists and that stale rule directories are gone) instead of a prose checklist of intentions.

State an explicit feedback loop for the install plan, e.g. 'if verification fails, fix the install and re-verify before reporting', since the workflow involves batch installs/removals of hooks and rules.

Lean out peripheral detail — the worked evidence-table example and the handoffs catalog — into a short reference file or trim them to one line each to reduce token cost of the always-loaded body.

Dimension	Reasoning	Score
Conciseness	The body is almost entirely dense, bulleted, non-obvious guidance (evidence commands, bucket criteria, table format) with no explanation of concepts Claude already knows. It sits at anchor 4 rather than 5 because sections like 'Outputs'/'Output Format' and the extended evidence-table example carry some redundancy that could be trimmed.	4 / 5
Actionability	Concrete, executable material is provided throughout: copy-paste ripgrep/cat commands, an exact evidence-table format, concrete install paths ('.claude/skills/skill-library/SKILL.md'), and a fixed output template. It stops short of anchor 5 because the verification step is a checklist of intentions ('verify every DAILY file exists where expected') without executable checks to run.	4 / 5
Workflow Clarity	A clear six-step sequence with an explicit step-6 verification checkpoint and a compact report including 'open questions', plus a defined pass split for parallel or sequential execution. The batch install/removal work has validation present (so the cap of 3 does not apply), but no explicit validate-fix-retry feedback loop, which keeps it at anchor 4 rather than 5.	4 / 5
Progressive Disclosure	The skill is well-sectioned with clear headers, and there is no inlined reference material that clearly belongs in separate files; no bundle files exist to reference. It is not anchor 5 because the body is a single ~215-line file that the under-50-line simple-skill exception does not cover, and peripheral detail (the worked evidence-table example, the handoffs catalog) could be leaned out into references.	4 / 5
	Total	16 / 20 Passed

Description

88%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 description that clearly states comprehensive concrete actions and an explicit 'Use when' clause with natural trigger phrasing. The only soft spot is modest keyword coverage and slight overlap with sibling trimming/cleanup skills.

Dimension	Reasoning	Score
Specificity	The description enumerates comprehensive concrete actions: 'build an evidence-backed ECC install plan', 'sorting skills, commands, rules, hooks, and extras into DAILY vs LIBRARY buckets using parallel repo-aware review passes'. It matches the level-5 anchor of multiple specific concrete actions with full component coverage, and exceeds anchor 4 because nothing significant is left implicit.	5 / 5
Completeness	It explicitly answers both questions: the 'what' (building an evidence-backed install plan by sorting components into DAILY vs LIBRARY buckets) and a concrete 'when' clause ('Use when ECC should be trimmed to what a project actually needs instead of loading the full bundle'). This matches anchor 5; it is not anchor 4 because the 'when' is explicit rather than merely present-but-imprecise.	5 / 5
Trigger Term Quality	Natural phrases like 'trimmed to what a project actually needs', 'full bundle', 'install plan', and 'repo' are phrases a user would plausibly say. It falls short of anchor 5 because common variations such as 'subset install', 'selective install', 'curate', or 'too noisy' are missing, though it is clearly above anchor 3 since several relevant natural terms are present.	4 / 5
Distinctiveness Conflict Risk	The ECC-sorting/DAILY-vs-LIBRARY niche is distinct with a clear trigger, but 'trimmed to what a project actually needs' has minor overlap risk with closely related context-trimming and catalog-cleanup skills (e.g. strategic-compact, skill-stocktake). It fits anchor 4 — mostly distinct with minor overlap — rather than anchor 5, whose example shows essentially zero conflict surface.	4 / 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
metadata_version	'metadata.version' is missing	Warning

	Total	15 / 16 Passed

Repository: affaan-m/ECC
Path: skills/agent-sort/SKILL.md
Commit: c70874f

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.