CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

create-rfc

Create OpenShell RFC proposals in rfc/ from a design request. Use when the user asks to write, draft, start, create, or update an RFC, Request for Comments, architecture proposal, API proposal, process proposal, or cross-cutting design proposal that should follow the OpenShell RFC process and template.

Quality

95%

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

SKILL.md
Quality
Evals
Security

Quality

Content

100%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.

An exemplary instruction-only skill body: lean, concrete, and well-sequenced. It assumes Claude's competence, delegates all detail to the repo's own RFC process and template documents, includes a validation checklist, and defines an explicit feedback loop for blocking unknowns.

DimensionReasoningScore

Conciseness

The ~47-line body is lean: it never explains concepts Claude already knows, and every line gives repo-specific instruction (paths, numbering, front-matter fields, section-level heuristics, boundaries like "Do not update published docs just because an RFC was drafted"). It matches the 5 anchor's 'every token earns its place'; the slight echo between step 8 and the Validation section is a checklist elaboration, not padding that would justify the 4 anchor.

5 / 5

Actionability

Concrete, executable guidance throughout: exact paths (rfc/README.md, rfc/0000-template/README.md, rfc/NNNN-short-title/README.md), exact rules (choose next NNNN, state: draft, hyphenated folder title, GitHub-handle fallback for author), and a conditional lint check. Per the rubric's scoring note, an instruction-only skill needs no code when guidance is this specific — it fully matches the 'copy-paste ready' spirit of the 5 anchor.

5 / 5

Workflow Clarity

The 8-step sequence is clearly ordered (read process docs → read template → pick number → create folder → fill front matter → draft sections → preserve open questions → re-check), with a dedicated Validation checklist before handoff and an explicit feedback loop ("If a missing decision blocks a coherent RFC, ask the user for that decision"). This matches the 5 anchor's explicit validation, error-recovery loop, and checklist; the operation is not destructive or batch, so no cap applies.

5 / 5

Progressive Disclosure

No bundle files exist, and none are needed: the body is under 50 lines and well-organized into Workflow, Writing Standards, and Validation sections, while delegating all template/process detail one level deep to rfc/README.md and rfc/0000-template/README.md instead of inlining it. Per the rubric's guideline for short, self-contained skills, this matches the 5 anchor.

5 / 5

Total

20

/

20

Passed

Description

91%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 that explicitly states what the skill does and when to use it, with rich, natural trigger synonyms. The only weaknesses are minor: the what-clause compresses the capability into one composite action, and generic proposal terms create slight overlap risk with non-RFC design-document requests.

DimensionReasoningScore

Specificity

"Create OpenShell RFC proposals in rfc/ from a design request" names the domain, input, and output location, with several action verbs (write, draft, start, create, update) appearing in the when-clause. It falls just short of the 5 anchor because the what-clause describes a single composite action rather than enumerating multiple distinct concrete capabilities, and it is more comprehensive than the 3 anchor's '1-2 concrete actions, not comprehensive'.

4 / 5

Completeness

It explicitly answers both what ("Create OpenShell RFC proposals in rfc/ from a design request") and when ("Use when the user asks to write, draft, start, create, or update an RFC...") with concrete trigger phrases, structurally matching the rubric's 5-anchor example. Not the 4 anchor because the when-clause is fully explicit and trigger-rich, not merely adequate.

5 / 5

Trigger Term Quality

The trigger list "write, draft, start, create, or update an RFC, Request for Comments, architecture proposal, API proposal, process proposal, or cross-cutting design proposal" covers natural verb synonyms and object synonyms users would actually say. This matches the 5 anchor's comprehensive synonym coverage; there are no file-extension variants applicable to RFC directories, so nothing natural is missing.

5 / 5

Distinctiveness Conflict Risk

The niche is clear (OpenShell RFC process, rfc/ directory, template), but broad trigger objects like "architecture proposal" and "cross-cutting design proposal" create minor overlap risk with a generic design-document skill when a user wants a proposal outside the RFC process. The qualifying clause "that should follow the OpenShell RFC process and template" mitigates this, keeping it above the 3 anchor (somewhat specific but overlapping) but short of the 5 anchor's minimal conflict risk.

4 / 5

Total

18

/

20

Passed

Validation

87%

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

Validation — 14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

metadata_field

'metadata' should map string keys to string values

Warning

Total

14

/

16

Passed

Repository

NVIDIA/OpenShell

Path

.agents/skills/create-rfc/SKILL.md

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.