claude-code-usage

[AUTO-INVOKE] MUST be invoked at the START of each new coding session. Covers context management, task strategies, and Foundry-specific workflows. Trigger: beginning of any new conversation or coding session in a Solidity/Foundry project.

Quality

45%

Does it follow best practices?

Impact

—

No eval scenarios have been run

Securityby

Passed

No known issues

Optimize this skill with Tessl

npx tessl skill review --optimize ./data/skills-md/0xlayerghost/solidity-agent-kit/claude-code-usage/SKILL.md

Quality

Content

50%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

This skill provides useful Foundry-specific commands and git safety rules, but is significantly weakened by mixing user-facing advice (meta-prompting, prompt techniques) with Claude-actionable instructions. The meta-prompting section, while well-structured, consumes substantial tokens on content that teaches the user how to prompt rather than teaching Claude how to act. The Foundry workflow section is the strongest part with concrete, executable commands.

Suggestions

Remove or drastically reduce the meta-prompting section — it's user-facing advice about how to prompt Claude, not instructions for Claude to follow. If retained, move it to a separate reference file.

Remove the 'Prompt Techniques' and 'Language Rule' sections — these describe user behavior and basic Claude capabilities, not project-specific actionable guidance.

Add validation checkpoints to the multi-file change workflow (e.g., 'after each file modification, run forge build to verify compilation before proceeding').

Focus the skill on what Claude should DO at session start: specific checks to run, files to read, project state to verify — rather than general advice about context management.

Dimension	Reasoning	Score
Conciseness	The skill is reasonably organized with tables for efficiency, but includes unnecessary content like the meta-prompting section which is extremely verbose with a full multi-round example in Chinese. The language rule and some prompt technique advice (avoid vague requests) are things Claude already knows. The meta-prompting section alone consumes ~40% of the document for what is essentially user-facing advice, not Claude-actionable instructions.	2 / 3
Actionability	The Foundry commands are concrete and executable (forge fmt, forge test -vvvv, etc.), and git rules are specific. However, much of the content is advisory/strategic rather than actionable by Claude — context management rules like 'use /clear over /compact' and meta-prompting templates are user-facing advice, not instructions Claude can execute. The skill mixes user guidance with Claude instructions.	2 / 3
Workflow Clarity	The task execution strategy table provides a reasonable decision tree, and the Foundry workflow has a clear sequence (build → fmt → test → gas-report). However, there are no explicit validation checkpoints or error recovery loops for the multi-file change workflow or the git operations. The meta-prompting flow is well-sequenced but is user-facing, not a Claude workflow.	2 / 3
Progressive Disclosure	The content uses tables and headers for organization, which helps navigation. However, the meta-prompting section with its full practical example is a large inline block that could be a separate reference file. There are no references to external files, and the document is somewhat monolithic at ~120 lines with no bundle files to offload detail. The mention of .claude/instructions.md is helpful but is the only external reference.	2 / 3
	Total	8 / 12 Passed

Description

40%

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 communicates when it should be triggered (start of a Solidity/Foundry coding session) but fails to describe what it actually does in concrete terms. The capabilities listed are abstract categories rather than specific actions, making it difficult for Claude to understand the skill's value or for it to be distinguished from other potential Foundry-related skills.

Suggestions

Replace vague categories like 'context management, task strategies, and Foundry-specific workflows' with concrete actions, e.g., 'Loads project structure, identifies existing contracts and test files, sets up Foundry build/test commands, and establishes coding conventions.'

Add specific trigger terms users might naturally use, such as 'new project', 'getting started', 'forge', 'foundry.toml', 'smart contracts'.

Clarify the 'what' by listing 2-4 specific outputs or behaviors the skill produces, e.g., 'Scans for foundry.toml configuration, maps contract dependencies, and summarizes test coverage status.'

Dimension	Reasoning	Score
Specificity	The description uses vague language like 'context management, task strategies, and Foundry-specific workflows' without listing any concrete actions. No specific capabilities are described—just abstract categories.	1 / 3
Completeness	The 'when' is clearly stated (beginning of any new conversation or coding session in a Solidity/Foundry project), but the 'what' is extremely vague—'context management, task strategies, and Foundry-specific workflows' doesn't explain what the skill actually does.	2 / 3
Trigger Term Quality	Includes some relevant keywords like 'Solidity', 'Foundry', 'coding session', and 'new conversation', but these are more about when to trigger than natural user language. Users wouldn't typically say 'context management' or 'task strategies'.	2 / 3
Distinctiveness Conflict Risk	The Solidity/Foundry niche and auto-invoke at session start provide some distinctiveness, but 'context management' and 'task strategies' are generic enough to potentially overlap with other project initialization or workflow skills.	2 / 3
	Total	7 / 12 Passed

Validation

100%

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 — 11 / 11 Passed

Validation for skill structure

No warnings or errors.

Repository: NeverSight/skills_feed
Commit: c7b7905

Reviewed: about 3 hours 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.