CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

pr-walkthrough

Create a narrated video walkthrough of a pull request with code slides and audio narration. Use when asked to create a PR walkthrough, PR video, or walkthrough video.

Quality

77%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./skills/pr-walkthrough/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

56%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.

The skill body is well-structured with a clear philosophy and mostly concrete, executable steps for the understand-write-narrate-audio pipeline. Its two real weaknesses are redundancy across the Philosophy, Script structure, and Narration tips sections, and a render step whose referenced tooling does not match the bundled scripts.

Suggestions

Fix the render step to match the actual bundle: replace ./video/render.sh with the real script (scripts/make-video.sh) and correct its documented inputs, or add the video/ tooling the body describes.

Explain how slide-NN.png images are produced (or reference a script that makes them), since the bundled make-video.sh requires them but no step covers slide image generation.

Merge the duplicated guidance: consolidate "explain the why" / "name files" advice into the Script structure section and fold API configuration into Step 3 rather than repeating it.

DimensionReasoningScore

Conciseness

The body is dense with real specifics, but key guidance is repeated: "Explain the why, not just the what" and "name files/functions" each appear three times (Philosophy, Code walkthrough guidelines, Narration writing tips), the File organization section restates the tmp/out layout already given up top, and API configuration re-states the GEMINI_API_KEY/voice details from Step 3.

3 / 5

Actionability

Steps 1-4 give executable commands and a complete narration.json example, but Step 5 instructs running "./video/render.sh" which does not exist in the bundle (the actual script is scripts/make-video.sh, taking slide-NN.png files the skill never explains how to produce), so the final, most fragile step is not executable as written.

3 / 5

Workflow Clarity

A clear 5-step sequence with stated dependencies, validation built into audio generation (duration sanity checks with automatic retries), and an end checklist including output verification. Minor gaps: no checkpoint for slide/manifest correctness before the render step, and the render step itself is misdocumented.

4 / 5

Progressive Disclosure

Sectioning is good and references/manifest.md is a real, well-signaled one-level reference, but the body and its file-organization tree reference an entire video/ directory (render.sh, build.mjs, hyperframes.json) that is absent from the actual bundle, leaving the reader navigating to files that do not exist.

3 / 5

Total

13

/

20

Passed

Description

87%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: concrete, third-person, and equipped with an explicit Use-when clause covering natural trigger phrases. Its only weakness is modest keyword coverage relative to the top anchor, where a few plausible synonyms are absent.

Suggestions

Add a couple of natural trigger variations, e.g. "record a walkthrough", "video summary of a PR", or "pull request video".

Consider mentioning the input form ("given a GitHub pull request URL") in the description so the trigger surface matches the skill's stated input.

DimensionReasoningScore

Specificity

The description names a concrete deliverable ("narrated video walkthrough of a pull request") with specific artifacts ("code slides and audio narration"), but describes a single composite action rather than the multiple distinct actions of the top anchor.

4 / 5

Completeness

It explicitly answers both questions: the "what" ("Create a narrated video walkthrough of a pull request with code slides and audio narration") and the "when" via a concrete "Use when asked to create a PR walkthrough, PR video, or walkthrough video" trigger clause.

5 / 5

Trigger Term Quality

"PR walkthrough, PR video, walkthrough video" are natural phrases a user would say, but common variations like "pull request video", "screencast", or "record a walkthrough of this PR" are missing.

4 / 5

Distinctiveness Conflict Risk

It occupies a clear niche (PR walkthrough videos) with distinct trigger phrases that no adjacent skill (code review, video editing, general narration) would naturally claim.

5 / 5

Total

18

/

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

tldraw/tldraw

Path

skills/pr-walkthrough/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.