CtrlK
BlogDocsLog inGet started
Tessl Logo

pantheon-ai/create-context-file

Create context files (plans, justifications, scratches) in .context/ directory with unique three-word IDs and frontmatter

68

Quality

86%

Does it follow best practices?

Impact

No eval scenarios have been run

SecuritybySnyk

Passed

No known issues

Overview
Quality
Evals
Security
Files

Quality

Content

65%

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 with concrete commands and examples, but suffers from cross-section redundancy, an implicit rather than explicit workflow, and an empty References section paired with a missing script bundle file. Tightening the duplicated file-type descriptions and filling/removing the References section would lift the weaker dimensions.

Suggestions

Consolidate the file-type descriptions: keep one of "What I do", "When to Use", or the "File Type Selection Guide" table and have the others reference it, removing the redundancy that hurts conciseness.

Promote the "ls .context/ first" check into a short numbered workflow (1. check existing, 2. run script, 3. verify created file) so validation checkpoints are explicit rather than hidden in an anti-pattern.

Either populate the "## References" section with the actual script path or remove the empty heading, and ensure the referenced create-context-file.ts script is included as a bundle file under scripts/.

DimensionReasoningScore

Conciseness

The three file types are described three times — in "What I do", "When to Use", and the "File Type Selection Guide" table — and the four anti-patterns each carry WHY/BAD/GOOD prose, so the body is mostly efficient but includes redundancy that could be tightened; it does not, however, explain concepts Claude already knows, so it stays above the bottom anchor.

2 / 3

Actionability

Provides fully executable commands ("bunx .context/skills/create-context-file/scripts/create-context-file.ts --type plan \"feature-name\" \"Plan content here\""), a heredoc form, and a concrete filename/frontmatter example — copy-paste ready and specific.

3 / 3

Workflow Clarity

The recommended pre-check ("ls .context/plans/ first") exists but is buried inside an anti-pattern rather than a numbered sequence, and there is no post-creation verification step, so checkpoints are implicit; not the bottom anchor because the single action is itself unambiguous.

2 / 3

Progressive Disclosure

Sections are well organized, but the "## References" heading is empty and the referenced script (.context/skills/create-context-file/scripts/create-context-file.ts) has no corresponding bundle file, so navigation/structure is incomplete rather than a clean one-level-deep disclosure.

2 / 3

Total

9

/

12

Passed

Description

100%

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 across all dimensions: it states concrete capabilities, provides explicit "Use when" triggers with natural user phrasing, and defines a recognizable niche. It is somewhat verbose with three elaborated when-clauses, but every clause is informative rather than padded.

DimensionReasoningScore

Specificity

Lists multiple concrete actions — "Creates structured context files (plans, justifications, scratches)", "unique three-word IDs", and "frontmatter metadata" — matching the anchor for listing several specific concrete actions rather than vague language.

3 / 3

Completeness

Explicitly answers both what ("Creates structured context files...") and when ("Use when starting a new task...") with an explicit "Use when" clause and concrete triggers, matching the top anchor.

3 / 3

Trigger Term Quality

Quotes natural phrases users would say — "create a plan", "document my thinking", "save context", "track decisions", "keep notes" — giving good coverage of real trigger language rather than technical jargon.

3 / 3

Distinctiveness Conflict Risk

The .context/ directory convention, three-word IDs, and the plan/justification/scratch trichotomy carve a clear niche; trigger words are generic individually but are scoped to a "structured project context directory", keeping conflict risk low.

3 / 3

Total

12

/

12

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.

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Reviewed

Table of Contents