CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

page-agent

Embed an in-page natural-language GUI copilot in web apps.

60

Quality

75%

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 ./optional-skills/web-development/page-agent/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

88%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.

This is a strong, highly actionable skill body: three well-sequenced adoption paths with executable code for each, per-path verification checkpoints, and a pitfalls section grounded in verified behavior. The main weaknesses are minor token waste from version commentary and inline repo-detail that could live in a reference file.

Suggestions

Trim version-specific commentary ("17k+ stars", "docs claim 11+ but 10.9 works fine", the npm 10-vs-11 note) or consolidate the version constraints into the Prerequisites section once.

Move the repo-layout table and browser-extension development details into a references/ file (e.g. REPO_LAYOUT.md) referenced one level deep, keeping SKILL.md as a lean overview.

Deduplicate the two bookmarklet code blocks (Path 1 CDN vs Path 3 local) into a single reusable snippet parameterized by script URL.

DimensionReasoningScore

Conciseness

The body is dense with non-obvious operational facts (engine constraints, CSP and CORS pitfalls, the verified .env key-inlining warning) rather than concepts Claude already knows, but a few tokens could be trimmed: "17k+ stars", the duplicated bookmarklet forms, and version commentary like "docs claim 11+ but 10.9 works fine". This matches 'efficient; minor instances of over-explanation' rather than the every-token-earns-its-place level.

4 / 5

Actionability

Every path is copy-paste ready: a CDN script tag and bookmarklet, `npm install page-agent` with a complete PageAgent constructor example, a provider table with real baseURL/model values, .env templates, and exact npm commands. This matches the fully executable, common-cases-covered anchor.

5 / 5

Workflow Clarity

Three clearly sequenced paths (demo / production / source) are each followed by an explicit 'Verifying it works' checklist (panel appears, request visible in Network tab, curl -I returns 200), and the Pitfalls section provides error-recovery loops for the common failure modes (CORS, bad baseURL, engine errors, restarting Vite after .env changes).

5 / 5

Progressive Disclosure

No bundle files exist, and the single-file body is well organized with clear section headers for each path, correctly deferring 'the full option list' to the external docs site. Minor gaps: the repo-layout table and extension-development detail are inline where a one-level-deep reference file could carry them, so it sits at 'good structure; minor organization gaps' rather than the well-split 5.

4 / 5

Total

18

/

20

Passed

Description

53%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.

The description is concise, in third person, and identifies a distinctive niche, but it is a single sentence that undersells the skill. It states one action with no 'when to use' trigger guidance and misses the natural synonyms users would say when they need this skill.

Suggestions

Add a 'Use when...' clause, e.g. 'Use when a user wants to add an AI copilot or natural-language interface to their own web app, dashboard, or admin panel.'

Include the core capability verbs and natural synonyms: executes natural-language instructions against the page DOM (clicking, filling forms), browser agent, AI assistant, page automation, page-agent.

Mention the key constraint that differentiates it from multimodal browser agents — text-DOM only, no screenshots, pure client-side — so it doesn't trigger for screenshot-based automation requests.

DimensionReasoningScore

Specificity

"Embed an in-page natural-language GUI copilot in web apps" names the domain and one concrete action (embedding a copilot), but omits what the copilot actually does (executes natural-language instructions against the DOM, no screenshots) — matching the anchor for 1-2 concrete actions, not the several specific actions required for 4.

3 / 5

Completeness

The 'what' is clear (embed an in-page NL GUI copilot in web apps) but there is no 'Use when...' clause or equivalent trigger guidance, which caps completeness at 3 per the judging guidelines.

3 / 5

Trigger Term Quality

Keywords present are "in-page", "natural-language", "GUI copilot", "web apps", but common phrases a user would naturally say — "browser agent", "AI assistant in my app", "page automation", "page-agent" — are missing, matching the 'some relevant keywords but missing common variations' anchor.

3 / 5

Distinctiveness Conflict Risk

"In-page GUI copilot" carves a clear niche distinct from document or data skills, but the description doesn't disambiguate from browser-automation requests (driving a browser from outside), leaving minor overlap risk with closely related skills.

4 / 5

Total

13

/

20

Passed

Validation

81%

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

Validation — 13 / 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

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

13

/

16

Passed

Repository

NousResearch/hermes-agent

Path

optional-skills/web-development/page-agent/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.