CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

create-prd

Draft PRD with scope, stories, acceptance criteria, verification. Use when shaping a new feature or spec.

54

Quality

68%

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

Fix and improve this skill with Tessl

tessl review fix ./.agents/skills/02-shape/create-prd/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

52%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-sequenced drafting process with a strong verification checklist, but it depends entirely on a template file (assets/prd-template.md) that is absent from the bundle, and it carries verbatim duplication of the shaping-file list and output paths. The structure and rubric content are genuinely useful, yet the broken asset reference leaves the skill not fully executable as shipped.

Suggestions

Ship the referenced assets/prd-template.md (or inline the minimum PRD structure) so Quick Start step 1 and the "Minimum PRD structure" section are self-contained.

Remove the duplicated shaping-file list ("Try find this information:" repeats the four files already listed under "No Context Loss") and consolidate the multi-PRD output rules stated in Quick Start, Outputs, and Steps.

Add a brief feedback loop under Verification telling the agent what to do when a check fails (e.g., revise the offending PRD section, re-split an oversized slice), which would lift workflow clarity toward the top anchor.

DimensionReasoningScore

Conciseness

The body is dense and checklist-like with no over-explanation of known concepts, but the shaping-file list (brief.md, breadboard-pack.md, spike-investigation.md, risk-register.md) is duplicated verbatim within one section ("No Context Loss" followed by "Try find this information:"), and the multi-PRD location rules are repeated three times (Quick Start, Outputs, Steps). This matches anchor 3 (mostly efficient but could be tightened); it exceeds anchor 2 but falls short of anchor 4, where only minor trims would be needed.

3 / 5

Actionability

The guidance is specific (concrete clarifying questions, section checklists, exact naming conventions like prds/<slice_id>_<slug>/prd.md), but the central executable step "Copy the template: assets/prd-template.md" points to a file that does not exist in the bundle, and the minimum structure is "defined by" that missing file. This matches anchor 3 (concrete guidance but incomplete, missing key details); not 4 because a key dependency is unavailable to execute.

3 / 5

Workflow Clarity

Five sequenced steps (clarify, run slicing rubric, template, fill sections, verify) are followed by an explicit "## Verification" checklist with about ten checkpoints, matching anchor 4 (clear sequence, most checkpoints present). It falls short of anchor 5, which requires feedback loops for error recovery (what to do when a verification check fails), and exceeds anchor 3, which only requires a sequence with implicit checkpoints.

4 / 5

Progressive Disclosure

Scored against the actual bundle: no references/, scripts/, or assets/ files exist at all, yet the body routes to assets/prd-template.md three times as the source of the required structure — a broken, dead-end reference. This matches anchor 2 (minimal effective structure / broken navigation); it does not merit anchor 3, because the missing file is the load-bearing reference rather than a minor organization gap, and the inlined slicing rubric plus repeated output rules could have lived in a separate file.

2 / 5

Total

12

/

20

Passed

Description

70%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 compact, third-person description with an explicit "Use when" clause and a clearly stated deliverable. It is lean and domain-specific, but its trigger coverage is limited to one when-condition and misses common phrasings like "product requirements document" or "write a PRD".

Suggestions

Add trigger variations users naturally say, e.g., "Use when the user asks to write a PRD, product requirements document, or feature spec, or when shaping a new feature before implementation".

List the concrete actions performed rather than one verb with components, e.g., "Draft PRDs with scope, user stories, numbered acceptance criteria, and verification plans; slice oversized scope into thin PRDs".

DimensionReasoningScore

Specificity

"Draft PRD with scope, stories, acceptance criteria, verification" names the domain and enumerates the deliverable parts, but relies on a single action verb ("Draft") rather than several distinct concrete actions, matching anchor 3. It is not anchor 4, which requires multiple specific actions (e.g., "extracts, fills, converts"), and not anchor 2 because the language is concrete rather than generic.

3 / 5

Completeness

Both parts are explicit: what ("Draft PRD with scope, stories, acceptance criteria, verification") and when ("Use when shaping a new feature or spec"). The single when-clause is narrower than anchor 5, which demands concrete multi-condition trigger phrases, but broader than anchor 3, where the when is missing or only implied.

4 / 5

Trigger Term Quality

Natural user terms like "PRD", "spec", "new feature", and "acceptance criteria" are present, but common variations such as "product requirements document", "requirements", or "write a PRD" are missing, matching anchor 4 (good coverage, a few natural terms missing). Not 5 because synonym and full-name coverage is absent; not 3 because the included terms are natural phrases users would say, not just jargon.

4 / 5

Distinctiveness Conflict Risk

PRD drafting is a clear niche, but "spec" and "shaping a new feature" have minor overlap with general planning or spec-writing skills, matching anchor 4 (mostly distinct; minor overlap risk). Not 5 because "spec" is a broad trigger that could pull in related skills.

4 / 5

Total

15

/

20

Passed

Validation

93%

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

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

referenced_paths_exist

Referenced path issues: 3 missing

Warning

Total

15

/

16

Passed

Repository

marchatton/agent-skills

Path

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