gws-gmail

Read and manage Gmail across all connected Google accounts — search, send, reply, draft, archive, and label emails.

Quality

85%

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

87%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A compact, well-structured reference with concrete executable MCP calls and clear account-routing rules, weakened only by missing validation/confirmation steps for irreversible write operations like sending email.

Suggestions

Add a pre-send validation checkpoint to the workflow (e.g. confirm recipients/subject/body with the user before sending, and verify the send succeeded) to lift workflow clarity.

Note what a successful send/reply response looks like so Claude can confirm the operation completed rather than assuming success.

Dimension	Reasoning	Score
Conciseness	The body is lean and assumes Claude's competence — subcommand/flag tables and short examples with no padding or explanation of what Gmail is.	3 / 3
Actionability	Provides concrete, copy-paste-ready MCP invocations like google_gmail(command: "list --query 'is:unread'") and a flags table with specifics such as the --max default of 20.	3 / 3
Workflow Clarity	The reading/replying workflow is sequenced and account routing has an ask-first checkpoint, but outward-facing write operations (send/reply) lack explicit validation/confirmation steps, capping this at 2 per the destructive-operations guideline.	2 / 3
Progressive Disclosure	A short, single-purpose skill with no bundle files; its well-organized sections (Overview, Subcommands, Account Routing, Key Flags, Workflow, Error Handling) satisfy the simple-skill bar for a 3.	3 / 3
	Total	11 / 12 Passed

Description

82%

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 concise, third-person description that names concrete Gmail actions and natural trigger terms, but it lacks an explicit 'Use when...' clause so completeness is capped at 2.

Suggestions

Append a 'Use when...' clause (e.g. 'Use when the user wants to search, send, reply to, draft, archive, or label Gmail messages') to explicitly answer when Claude should engage this skill.

Consider naming common user phrasings like 'check my inbox' or 'find that email' to broaden natural trigger coverage.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — "search, send, reply, draft, archive, and label" — matching the top anchor's bar of several specific verbs rather than a vague domain label.	3 / 3
Completeness	It clearly answers "what" (read/manage Gmail with enumerated actions) but offers no "Use when..." clause or equivalent explicit trigger guidance, so the when is only implied.	2 / 3
Trigger Term Quality	"Gmail", "search", "send", "reply", "draft", "archive", and "label" are natural terms users actually say, giving good coverage rather than jargon.	3 / 3
Distinctiveness Conflict Risk	The Gmail-specific framing and explicit action list carve a clear niche unlikely to trigger for other skills.	3 / 3
	Total	11 / 12 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: accomplish-ai/coworker
Path: apps/desktop/bundled-skills/gws-gmail/SKILL.md
Commit: 2cf74d0

Reviewed: 1 day 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.