decide-build-primitive

Analyze, compare, and recommend a Codex build primitive. Use when the user is packaging or automating a workflow and the right primitive is unclear.

Quality

74%

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

Fix and improve this skill with Tessl

tessl review fix ./Skills/agent-ops/decide-build-primitive/SKILL.md

Quality

Content

77%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 body is token-efficient and the workflow is cleanly sequenced with validation feedback loops, but it omits the actual decision criteria that make a routing skill actionable and its progressive-disclosure pointer references a path that does not exist instead of the real reference bundle.

Suggestions

Add a concrete decision matrix or criteria mapping inputs/autonomy/trigger frequency to Skill vs Custom Prompt vs Agent so the recommendation step is actionable rather than asserted.

Fix the Progressive Disclosure pointer to reference the real bundle (./references/contract.yaml, evals.yaml, task-profile.json) or move those files under the cited archive path so signaled references resolve.

Dimension	Reasoning	Score
Conciseness	The body is lean and well-sectioned with no padding or explanations of concepts Claude already knows; every section (Philosophy, Inputs, Workflow, Constraints) earns its place, matching the level-5 anchor.	5 / 5
Actionability	As an instruction-only decision skill code is not required, but the actual decision criteria for choosing Skill vs Custom Prompt vs Agent are missing, leaving key how-to-decide details incomplete per the level-3 anchor.	3 / 5
Workflow Clarity	The five-step workflow is clearly sequenced with explicit validation checkpoints ('Stop at the first failed gate', 'Rerun the relevant validation after fixes') and a feedback loop, matching the level-5 anchor; the destructive/batch cap does not apply to this routing skill.	5 / 5
Progressive Disclosure	Sections are well-organized, but the Progressive Disclosure pointer references a non-existent archive path (Infrastructure/references/deferred-skill-context/...) while the real bundle files in ./references/ (contract.yaml, evals.yaml, task-profile.json) are never signaled, so navigation is misdirected rather than clear.	3 / 5
	Total	16 / 20 Passed

Description

71%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 cleanly answers both what and when with an explicit trigger clause, but its action verbs are generic and the key primitive nouns (skill/prompt/agent) that users would naturally say are missing from the description.

Suggestions

Add the concrete decision target nouns ('Skill, Custom Prompt, or Agent') to the description so the natural trigger terms users say are present.

Replace abstract verbs with more concrete operations (e.g., 'compare Skill, Custom Prompt, and Agent options across inputs, autonomy, and trigger frequency') to lift specificity toward level 4-5.

Dimension	Reasoning	Score
Specificity	Names the domain ('Codex build primitive') and three actions ('Analyze, compare, and recommend'), but the verbs are abstract rather than concrete operations like the level-4 anchors' 'extracts/fills/converts', so it sits above 2 yet below 4.	3 / 5
Completeness	It explicitly states what ('Analyze, compare, and recommend a Codex build primitive') and when ('Use when the user is packaging or automating a workflow and the right primitive is unclear'), matching the level-5 anchor with a concrete trigger clause.	5 / 5
Trigger Term Quality	'packaging or automating a workflow' are natural phrases, but the core decision nouns (skill, prompt, agent) and synonyms are absent and 'Codex build primitive' leans toward jargon, leaving common variations missing.	3 / 5
Distinctiveness Conflict Risk	'Codex build primitive' is a distinct niche with minimal conflict risk, but 'packaging or automating a workflow' is broad enough to overlap with general automation skills, keeping it just below 5.	4 / 5
	Total	15 / 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: jscraik/Agent-Skills
Path: Skills/agent-ops/decide-build-primitive/SKILL.md
Commit: f1f2f21

Reviewed: 2 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.