CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

inspecting-hermes-desktop-dom

Read the live Hermes desktop DOM/CSS over CDP.

Quality

74%

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/software-development/inspecting-hermes-desktop-dom/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 operational skill: fully executable commands with explicit validation checkpoints and error-recovery pitfalls, written for a competent reader with almost no padding. The only weaknesses are minor — a couple of trimmable asides and a stale-repo-note that will rot. Structure is appropriate for a self-contained single-file skill.

DimensionReasoningScore

Conciseness

The body is dense and assumes Claude's competence (CDP, Electron single-instance lock, Vite dev mode) with no basic-concept padding, but a few tokens could be trimmed: "The renderer is a Chromium page, so everything DevTools can read, a script can read" and the time-sensitive stale-note about profile-typing-lag.md, which is repo-state commentary not placed in an old-patterns/deprecated section. Fits anchor 4 (efficient, minor trimmable instances) rather than 5.

4 / 5

Actionability

Every path is copy-paste ready: the curl port check with --max-time and env-var default, the `node scripts/eval.mjs "..."` one-liner, a complete CDP.connect import snippet with match and eval usage, and a full isolated-instance launch command with HERMES_HOME, port, and --user-data-dir. This matches anchor 5 (fully executable commands covering the common cases).

5 / 5

Workflow Clarity

The flow is explicit and validated: check the port first ("Check before doing anything else"), eval against the user's window or launch an isolated instance when no port exists, with concrete validation checkpoints (curl response, "DevTools listening on ws://... — that line is the proof the port opened", "Poll, don't probe once") and error-recovery guidance for ECONNREFUSED and self-exited probes. Operations are read-only, so the destructive/batch cap does not apply; this is anchor 5.

5 / 5

Progressive Disclosure

No bundle files exist, and the body is well-sectioned (Overview / When to Use / The port / Reading the DOM / isolated instance / Pitfalls) while deferring bulk detail to real repo files (eval.mjs, scripts/perf/lib/cdp.mjs) instead of inlining them. It sits at anchor 4 rather than 5 because at ~145 lines, with the under-50-line simple-skill exception unavailable, sections like the winning-rule deep dive could arguably live in a separate reference file.

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 and unambiguous about what the skill does, with a distinctive niche, but it omits any 'use when' trigger guidance and lists only a single action. It reads as a capability label rather than an invitable trigger. Adding trigger phrases and one or two more concrete actions would lift it substantially.

Suggestions

Add an explicit trigger clause, e.g. "Use when a UI change in apps/desktop isn't showing up in the running app, when you need to check which CSS rule won, or to read renderer console errors the user can't copy out."

Mention the concrete read operations the body covers — computed styles, geometry, winning-rule lookup, console output — so the description's capability list is comprehensive rather than a single 'read the DOM' action.

Include natural synonyms users would actually say ("DevTools", "inspect element", "computed style") and briefly disambiguate from perf/heap debugging skills to further reduce overlap risk.

DimensionReasoningScore

Specificity

"Read the live Hermes desktop DOM/CSS over CDP" names the domain (Hermes desktop, DOM/CSS, CDP) and one concrete action, but stops there — capabilities covered in the body (computed styles, winning CSS rule, console output) are absent. This matches anchor 3 (domain plus 1-2 concrete actions, not comprehensive), not 4, which requires several listed actions.

3 / 5

Completeness

The description has a clear 'what' ("Read the live Hermes desktop DOM/CSS over CDP") but no 'Use when...' clause or equivalent explicit trigger guidance, which caps completeness at 3 per the judging guidelines. Not 4 because the 'when' is entirely absent rather than merely imprecise.

3 / 5

Trigger Term Quality

"Hermes desktop", "DOM", "CSS", and "CDP" are relevant keywords, but common natural phrasages users would say — "computed styles", "DevTools", "inspect element", "why isn't my style applying" — are missing. Fits anchor 3 (some relevant keywords, missing variations/synonyms) rather than 4's good-but-gappy coverage.

3 / 5

Distinctiveness Conflict Risk

"Hermes desktop DOM/CSS over CDP" carves a clear niche (live UI inspection of one specific app) with distinct triggers, giving minimal conflict risk with generic skills. It stays at 4 rather than 5 because closely related skills (node-inspect-debugger, debugging-hermes-desktop) exist and the description does nothing to disambiguate against them.

4 / 5

Total

13

/

20

Passed

Validation

75%

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

Validation — 12 / 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

referenced_paths_exist

Referenced path issues: 2 missing, 1 deeper-than-1-level

Warning

Total

12

/

16

Passed

Repository

NousResearch/hermes-agent

Path

optional-skills/software-development/inspecting-hermes-desktop-dom/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.