ce-optimize

Run metric-driven optimization loops. Use when improving measurable outcomes such as search relevance, clustering quality, build performance, prompt quality, or scored behavior through experiments.

Quality

87%

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

85%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 body is highly actionable and exceptionally well-structured, with explicit validation checkpoints and well-organized external references. Its main weakness is conciseness: the durability/re-read-from-disk message is repeated across several sections.

Suggestions

Consolidate the repeated crash-safety rationale: state the 'context window is not durable / re-read from disk' rule once in Persistence Discipline and reference it from the per-phase checkpoints rather than restating it.

Trim the duplicated 'Why immediately + verify?' explanation in Phase 3.3 since it restates Persistence Discipline's core rules.

Consider formatting reference paths as markdown links (e.g. [usage-guide.md](references/usage-guide.md)) so navigation is clickable rather than bare code paths.

Dimension	Reasoning	Score
Conciseness	The body is procedural and does not explain concepts Claude already knows, but the crash-safety / 're-read from disk' rationale is restated across Persistence Discipline, the CP checkpoints, Phase 3.3's 'Why immediately + verify?', and Phase 3.5, so it could be tightened.	2 / 3
Actionability	Provides concrete, executable bash commands with full argument signatures (measure.sh, parallel-probe.sh, experiment-worktree.sh), copy-paste-ready YAML examples (baseline, hypothesis_backlog, stratification, rubric), and a filled summary template.	3 / 3
Workflow Clarity	Phases 0–4 are clearly sequenced with explicit mandatory checkpoints (CP-0…CP-5), write-then-verify feedback loops, an error-recovery table, and crash-recovery steps — matching the score-3 anchor.	3 / 3
Progressive Disclosure	Schemas, templates, agent sub-prompts, and scripts are externalized to clearly signaled one-level-deep reference files, all of which exist in the bundle; the body retains only the orchestration logic.	3 / 3
	Total	11 / 12 Passed

Description

90%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 strong: it states the action, gives an explicit 'Use when' trigger, and lists concrete natural-language outcome domains. Its only weakness is specificity, since it names a single action rather than multiple concrete operations.

Dimension	Reasoning	Score
Specificity	States one concrete action ("Run metric-driven optimization loops") and lists example outcome domains (search relevance, clustering quality, build performance, prompt quality), but does not enumerate multiple concrete actions the way the score-3 anchor does.	2 / 3
Completeness	Explicitly answers both what ("Run metric-driven optimization loops") and when ("Use when improving measurable outcomes such as...") with an explicit trigger clause.	3 / 3
Trigger Term Quality	Covers several natural phrases a user would actually say ("search relevance", "clustering quality", "build performance", "prompt quality") and uses the natural "Use when improving measurable outcomes" trigger pattern.	3 / 3
Distinctiveness Conflict Risk	Occupies a clear niche (metric-driven iterative optimization through experiments) with distinct, outcome-specific triggers unlikely to fire for unrelated skills.	3 / 3
	Total	11 / 12 Passed

Validation

81%

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 — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (680 lines); consider splitting into references/ and linking	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
referenced_paths_exist	Referenced path issues: 2 deeper-than-1-level	Warning

	Total	13 / 16 Passed

Repository: EveryInc/compound-engineering-plugin
Path: skills/ce-optimize/SKILL.md
Commit: a9f6d53

Reviewed: 7 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.