CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

web-apps

Build an app that runs in the Open SWE web UI, preview it beside the thread with preview_app, and release it with publish_app. Read this before building an Open SWE app or plugin.

Quality

90%

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

The canonical home for this skill is web-apps in langchain-ai/open-swe

SKILL.md
Quality
Evals
Security

Quality

Content

93%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.

Exemplary content: dense with environment-specific facts, fully executable commands and examples, and a clean build-preview-publish loop. The single gap is the absence of an explicit validation checkpoint in the iteration workflow before publishing.

DimensionReasoningScore

Conciseness

Every sentence carries Open SWE-specific information Claude cannot know ("An app is one self-contained HTML file", "The frame's policy blocks every request, so a CDN script, a fetch or a font URL fails", the icon enum, permission semantics) and nothing explains concepts Claude already knows; there is no padding to trim.

5 / 5

Actionability

Guidance is copy-paste ready: a complete executable bash block (clone, sparse-checkout, npm install, build), a full valid manifest example, exact import paths and tokens ("@langchain/macaw-components/<Component>", "p-space-4"), and precise tool targets ("call preview_app with bundle/index.html"), with host API detail correctly deferred to the SDK's README.md and src/client.ts.

5 / 5

Workflow Clarity

The "Iterating" section gives a clear ordered loop (build → preview_app → change/rebuild → re-preview → publish_app) with a feedback channel ("The person sees it beside the thread and can chat with you from it"), but there is no explicit validation checkpoint such as confirming the build succeeded or the preview loaded before publishing — the 5 anchor's explicit validation steps are absent.

4 / 5

Progressive Disclosure

The ~57-line body is tightly organized into overview, manifest, starter setup, and iteration sections, with detail pushed one level deep to external SDK docs rather than inlined; no bundle files exist and none are needed, so well-organized sections with appropriately deferred detail meet the top anchor.

5 / 5

Total

19

/

20

Passed

Description

83%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: it names the domain, three concrete lifecycle actions with the exact tools that perform them, and an explicit read-before trigger. Its only weakness is a narrow trigger clause that covers 'building' but lacks the synonym and variation coverage of the best examples.

DimensionReasoningScore

Specificity

The description lists multiple concrete actions covering the full lifecycle — "Build an app that runs in the Open SWE web UI", "preview it beside the thread with preview_app", "release it with publish_app" — each tied to a named tool, with no lifecycle step missing.

5 / 5

Completeness

Both halves are answered: a clear 'what' (build, preview with preview_app, release with publish_app) and an explicit 'when' ("Read this before building an Open SWE app or plugin"). The trigger clause is a single narrow condition rather than the range of concrete trigger phrases the 5 anchor requires.

4 / 5

Trigger Term Quality

Natural terms a user would say are present ("app", "plugin", "Open SWE", "build"), but coverage stops there — no synonyms or trigger variations (e.g., previewing or publishing an existing app), matching the 'good coverage, a few natural terms missing' anchor rather than comprehensive synonym coverage.

4 / 5

Distinctiveness Conflict Risk

Distinctive markers ("Open SWE web UI", "preview_app", "publish_app", "Open SWE app or plugin") carve out a clear niche with tool-specific triggers, making conflict with other skills minimal.

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

langchain-ai/open-swe

Path

mda/bundled_skills/web-apps/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.