CtrlK
BlogDocsLog inGet started
Tessl Logo

llm-friendly-context

Clarifies inputs, outputs, success criteria, decisions, and unresolved conditions so downstream consumers can execute without guessing. Use when writing or revising LLM-facing prompts, handoffs, planning artifacts, reviews, reports, or generated instructions.

67

Quality

84%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

The canonical home for this skill is llm-friendly-context in shinpr/claude-code-workflows

SKILL.md
Quality
Evals
Security

Quality

Content

78%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 instruction-only skill: lean, free of filler, with a genuinely useful ambiguity-to-rewrite table and two concrete pre-flight checklists as validation checkpoints. The main gaps are a few cryptic, example-less directives (rules 6–7) and the lack of an explicit ordered apply-time workflow.

Suggestions

Tighten rules 6 and 7 into plain, checkable statements — e.g., rewrite the line-budget bullet as 'If the source said minimal/a few lines, cap the completed diff at that budget, counted once across all files' — and drop the duplicated prohibition checklist item from one of the two checklists.

Add one short before/after rewrite example (a vague instruction and its rewritten form) so the Rewrite Patterns table has a worked demonstration, not just row-level mappings.

Add a brief ordered 'How to apply' sequence (draft/rule check → rewrite table → matching checklist → send) so the rules, patterns, and checklists form an explicit workflow with the checklists as validation gates.

DimensionReasoningScore

Conciseness

The body is dense and lean with no padding or explanation of concepts Claude already knows; every section is rule or checklist content. It misses a 5 because rules 6 and 7 are wordy and near-cryptic (e.g., 'Apply `minimal`, `a few lines`, and explicit line estimates to the completed diff as one total budget') and the prohibition checklist item is duplicated verbatim in both checklists.

4 / 5

Actionability

The Rewrite Patterns table (9 ambiguous forms mapped to concrete rewrites) and the two checklists give largely executable, example-driven guidance, satisfying the instruction-only-skill standard. A few directives lack a before/after example — notably the line-budget rule and rule 6's uncertainty-handling bullets — leaving minor gaps that keep it below fully executable coverage of common cases.

4 / 5

Workflow Clarity

Both checklists act as explicit validation checkpoints before sending an artifact, and rule 5 requires a verification checkpoint per decomposed step. It falls short of a 5 because no explicit ordered sequence connects the rules, rewrite patterns, and checklists into an apply-time workflow, and there is no validate-fix-retry loop.

4 / 5

Progressive Disclosure

This is a single-file skill (~85 lines) with no bundle files and no need for external references, cleanly sectioned into Core Rules, Rewrite Patterns, Handoff Checklist, and Generated Artifact Checklist. Per the rubric's simple-skill note, well-organized sections with no nested references warrant the top score.

5 / 5

Total

17

/

20

Passed

Description

83%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: third-person, compact, with an explicit 'Use when' trigger clause and a clear statement of what the skill does. Its only weaknesses are a single action verb spread over many objects and slightly thin synonym coverage in the trigger terms.

DimensionReasoningScore

Specificity

"Clarifies inputs, outputs, success criteria, decisions, and unresolved conditions" names several concrete objects of the skill's action, matching the several-specific-actions anchor. It falls short of a 5 because there is only one distinct action verb rather than multiple comprehensive concrete actions.

4 / 5

Completeness

It explicitly answers both what ('Clarifies inputs, outputs, success criteria, decisions, and unresolved conditions so downstream consumers can execute without guessing') and when ('Use when writing or revising LLM-facing prompts, handoffs, planning artifacts, reviews, reports, or generated instructions') with concrete trigger phrases, matching the top anchor.

5 / 5

Trigger Term Quality

The 'Use when' clause lists natural terms users would say — 'LLM-facing prompts, handoffs, planning artifacts, reviews, reports, or generated instructions' — giving good keyword coverage. A few natural variants are missing (e.g., 'specs', 'context files', 'agent instructions'), so it does not reach comprehensive synonym coverage.

4 / 5

Distinctiveness Conflict Risk

'LLM-facing prompts', 'handoffs', and 'downstream consumers' carve a distinct niche for prompt/handoff authoring. The broad nouns 'reviews, reports' create minor overlap risk with generic writing or code-review skills, keeping it below the minimal-conflict anchor.

4 / 5

Total

17

/

20

Passed

Validation

100%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
shinpr/claude-code-workflows
Reviewed

Table of Contents

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.