desktop

Grida Desktop Electron shell and release-impact work: BrowserWindow, preload, `window.grida`, menus, protocol/deep links, file associations, Forge, path-scoped bridge security, Electron-only UI bugs, and CDP / Playwright verification. Use for `desktop/`, `editor/app/desktop/**`, `editor/scaffolds/desktop/**`, `editor/lib/desktop/**`, `/desktop/*` CSP, GRIDA-SEC-004, and deciding whether linked-package or hosted-renderer changes require a native Desktop version bump or coordinated release. For implementing daemon/agent-tenant core behavior, use `agent-system` as well.

Quality

94%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

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.

The body is a well-structured, highly actionable overview of the Grida Desktop Electron shell with executable commands, clear workflows, and explicit validation checkpoints. It is dense and mostly token-efficient, with only minor trimmable redundancy and a few sections that could be split into reference files.

Suggestions

Tighten the 'Boundaries' section: the 'Do not put OPFS...' and 'Do not add core behavior...' lines restate the four invariants and could be folded into invariant 4 to reduce redundancy (boosts conciseness).

Consider moving the full release-impact decision tree into a short reference file (e.g. references/release-impact.md) and summarizing the rule plus the audit script in SKILL.md, keeping deep navigation one level deep (boosts progressive_disclosure).

The Bridge section's two full TSX examples could be condensed to one canonical pattern with a one-line note for the soft-branch variant (boosts conciseness).

Dimension	Reasoning	Score
Conciseness	The body is dense and codebase-specific (paths, invariants, bridge contracts Claude does not already know), with little generic-concept padding. It is a 4 rather than 5 because some boundary prose ('Do not put OPFS...', 'Do not add core behavior...') partially restates the four invariants and could be tightened. Not a 3 because there is no over-explanation of concepts Claude already knows.	4 / 5
Actionability	Provides copy-paste-ready, executable guidance throughout — `pnpm --filter editor dev` / `pnpm --dir desktop dev`, `lsof -iTCP:9222`, CDP probe JS, Playwright `chromium.connectOverCDP`, the audit script, and `typecheck`/`test` owner checks. Not a 4 because commands and code cover the common cases completely with no gaps.	5 / 5
Workflow Clarity	Multi-step workflows are clearly sequenced with explicit validation checkpoints: two-terminal local setup, CDP/Playwright verification, a numbered Verification checklist with owner checks, and a release-impact decision tree that requires recording a status line. Not a 4 because validation steps (typecheck/test, audit, cold reload) are explicit rather than implicit.	5 / 5
Progressive Disclosure	Clear section headers, one-level-deep references to adjacent skills, CONTRIBUTING.md, SECURITY.md, a real bundled script (`scripts/audit-release-impact.sh`), and a Pointers index make navigation easy. It is a 4 rather than 5 because some inline material (the full release-impact decision tree, complete Bridge code examples) could plausibly live in a reference file, though it is appropriately overview-level. Not a 3 because structure and signaling are clearly present, not buried.	4 / 5
	Total	18 / 20 Passed

Description

100%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 a strong, third-person statement that comprehensively covers what the skill does and when to use it, with concrete path and identifier triggers. It also cleanly distinguishes itself from the adjacent agent-system skill, keeping conflict risk minimal.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions and surfaces — 'BrowserWindow, preload, `window.grida`, menus, protocol/deep links, file associations, Forge, path-scoped bridge security... CDP / Playwright verification' — giving comprehensive, non-abstract coverage. Not a 4 because the enumeration is broad and specific rather than having minor gaps.	5 / 5
Completeness	Explicitly answers 'what' ('Grida Desktop Electron shell and release-impact work...') and 'when' ('Use for `desktop/`... GRIDA-SEC-004, and deciding whether... require a native Desktop version bump or coordinated release') with concrete trigger phrases. Not a 4 because both what and when are explicit and specific, not merely present.	5 / 5
Trigger Term Quality	Natural dev-facing keywords ('Electron', 'preload', 'deep links', 'Forge', 'CDP / Playwright', 'version bump', 'release') plus concrete path triggers (`desktop/`, `editor/app/desktop/**`) and the GRIDA-SEC-004 identifier give comprehensive coverage including synonyms. Not a 4 because common variations and concrete identifiers are all present.	5 / 5
Distinctiveness Conflict Risk	Clear niche (Grida Desktop Electron shell) with distinct path/ID triggers, and it explicitly routes daemon/agent-tenant core behavior to `agent-system`, minimizing wrong-skill triggering. Not a 4 because the boundary against the sibling skill is explicitly stated.	5 / 5
	Total	20 / 20 Passed

Validation

87%

Warnings & errors only

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

Validation — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
relative_links	Relative link issues: 4 suspicious	Warning
referenced_paths_exist	Referenced path issues: 1 missing	Warning

	Total	14 / 16 Passed

Repository: gridaco/grida
Path: .agents/skills/desktop/SKILL.md
Commit: 8ef5e53

Reviewed: about 11 hours ago

Table of Contents

Discovery Implementation Validation

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.