CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

create-json-prd

Generate a Product Requirements Document (PRD) as JSON for Ralph by converting an existing PRD markdown file. Triggers on: create a prd, write prd for, plan this feature, requirements for, spec out.

64

Quality

81%

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

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 highly actionable, concrete instruction skill with a complete schema template, explicit output rules, and a verification checklist. Its weaknesses are modest repetition of the quality-gates emphasis, an unnumbered tail of workflow sections, and an orphaned example file in references/examples/ that the body never points to.

Suggestions

Reference the bundled example in the body, e.g., under Step 2: "See [json-prd-example.md](references/examples/json-prd-example.md) for a complete worked example" — this surfaces the existing asset and could let the inline template be trimmed.

State the error-recovery loop after Verification (e.g., "If any check fails, fix the JSON and re-verify before reporting") and number or merge the post-Step-2 sections so the full workflow reads as one sequence.

Consolidate the triple quality-gates emphasis (The Job step 3, the Step 1 focus bullet, and the example question) into one authoritative statement to save tokens.

DimensionReasoningScore

Conciseness

The body is prescriptive and assumes competence — it never explains what a PRD or JSON is — with exact paths, schemas, and commands. Minor trimmable repetition: quality gates are stated three times ("Always ask about quality gates", the "Quality Gates... (REQUIRED)" bullet, and the example question), keeping it below anchor 5 but well above anchor 3.

4 / 5

Actionability

Fully executable guidance throughout: exact input/output path mappings ("prd.md -> prd.json"), a complete copy-paste JSON schema template, a concrete question-format example, explicit install-command requirements ("pnpm add <pkg>"), and a verbatim closing message ("PRD JSON saved to <path>. Close this chat and run \`ralph build\`.\") with no gaps for the common cases.

5 / 5

Workflow Clarity

A clear sequence (The Job -> Step 1 clarifying questions -> Step 2 JSON structure), a NO-GO gate when no PRD markdown exists, and a concrete Verification checklist ("JSON parses", "contains required keys", "no loss"). Minor gaps: no explicit error-recovery loop after verification, and the post-Step-2 sections (No Information Loss, Output Requirements, Verification) are unnumbered, so the full flow is slightly less crisp than anchor 5.

4 / 5

Progressive Disclosure

The body is well-sectioned and the inline JSON schema template legitimately belongs inline, but the bundled references/examples/json-prd-example.md — a complete, ready-made example PRD JSON — is never referenced anywhere in the body, leaving a useful bundle file undiscoverable. That is more than the minor organization gap of anchor 4, though the body's own structure is far better than anchor 2's minimal/inlined-content characterization.

3 / 5

Total

16

/

20

Passed

Description

75%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 clear, well-scoped description that explicitly states what the skill does and when to trigger it, with natural trigger phrases. Its main weaknesses are narrow action coverage in the description itself and a trigger term ("create a prd") that overlaps with the sibling markdown-PRD creation skill.

Suggestions

Broaden the 'what' with 2-3 more concrete actions from the body (e.g., 'ask clarifying questions, capture quality gates, and preserve all PRD details with no information loss') to lift specificity.

Resolve the collision with the sibling $create-prd skill: replace or qualify the 'create a prd' trigger with conversion-oriented phrases like 'convert prd to json', 'prd.json', or 'json prd for ralph'.

DimensionReasoningScore

Specificity

"Generate a Product Requirements Document (PRD) as JSON for Ralph by converting an existing PRD markdown file" names the domain and one concrete conversion action, but coverage is not comprehensive (no mention of quality-gate elicitation, no-loss mapping, or output naming rules). Anchor 4 requires several distinct specific actions; this is a single action described two ways.

3 / 5

Completeness

Both questions are explicitly answered: the "what" (generate PRD JSON by converting an existing PRD markdown file) and the "when" (an explicit "Triggers on:" clause with five concrete trigger phrases, equivalent to a "Use when..." clause). Not anchor 4, since the "when" is already explicit and specific with concrete phrases.

5 / 5

Trigger Term Quality

"Triggers on: create a prd, write prd for, plan this feature, requirements for, spec out" gives good natural-phrase coverage users would actually say. A few natural terms are missing (e.g., "prd.json", "convert prd to json", "product requirements"), so it falls short of anchor 5's comprehensive synonym/extension coverage.

4 / 5

Distinctiveness Conflict Risk

The JSON-conversion niche is fairly distinct ("as JSON", "for Ralph"), but the trigger "create a prd" collides directly with the sibling $create-prd skill that writes the markdown PRD — a closely related skill — which is more than the minor overlap anchor 4 allows.

3 / 5

Total

15

/

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

marchatton/agent-skills

Path

.agents/skills/02-shape/create-json-prd/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.