google

Google Workspace CLI for Gmail, Calendar, Drive, Docs, Sheets, and Slides using gogcli. Use when asked to search email, schedule meetings, manage files, edit documents, update spreadsheets, or create presentations.

Quality

78%

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

Fix and improve this skill with Tessl

tessl review fix ./skills/google/SKILL.md

Quality

Content

68%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.

A well-organized CLI reference with excellent progressive disclosure into per-service files and nearly complete, executable command coverage. The main gaps are the absence of validation steps around destructive and batch operations (capping workflow clarity), a few non-executable placeholders, and an orphaned bundle file.

Suggestions

Add validation guidance to destructive and batch operations — e.g., before `gog drive delete`/`gog sheets clear`, verify the target with `gog drive get <fileId>` or `gog --dry-run drive delete <fileId>`, and confirm batch results after the download/export loop. This would lift the workflow_clarity cap.

Fix the executable gaps: show how to resolve the version in the manual-install tarball URL (e.g., fetch the release tag first), and align the error-table's `gog auth setup` with the authentication commands actually documented (`gog auth credentials set` / `gog auth add`).

Link `references/permissions.md` from SKILL.md (e.g., from the Agent Safety or Drive section) so it is discoverable, and drop or trim the Quick Reference block that duplicates commands already shown in the Commands sections.

Dimension	Reasoning	Score
Conciseness	The body is command-first with terse comments and almost no over-explanation ("All commands support `--json` for structured output and `--plain` for TSV"), assuming Claude's competence. Not 5 because the Quick Reference block duplicates commands restated verbatim in the Commands sections (e.g., `gog gmail search "is:unread"`, `gog drive search "quarterly report"`, `gog sheets get`, `gog slides info`), and the Examples section repeats `freebusy` from the Calendar section.	4 / 5
Actionability	Nearly every command is copy-paste ready with concrete arguments, and examples cover common cases end-to-end. Not 5 because of minor gaps: the manual-install URL contains an unresolved `gogcli_VERSION_linux_amd64.tar.gz` placeholder with no instruction for resolving the version, and the error-handling table cites `gog auth setup` while the Authentication section only shows `auth credentials set`/`auth add`/`auth doctor`.	4 / 5
Workflow Clarity	The auth sequence has an explicit validation checkpoint (`gog auth doctor`) and the exit-code table with retryability is a feedback loop, but the skill exposes destructive operations (`drive delete`, `gmail trash`, `calendar delete`, `sheets clear`) and a batch loop ("Download recent documents") with no verify-before/verify-after steps. Per the rubric, missing validation in destructive/batch contexts caps this at 3 — the `--dry-run` and `--readonly` flags exist in the Agent Safety section but are never wired into any destructive workflow.	3 / 5
Progressive Disclosure	The SKILL.md is a genuine overview with clearly signaled one-level-deep per-service references ("See [gmail.md](references/gmail.md) for the full command reference"), and the linked files all exist. Not 5 because scoring against the actual bundle reveals `references/permissions.md` (179 lines) is orphaned — linked from neither SKILL.md nor any reference file — and the Docs Workflows subsection uses a misplaced `####` heading level under `### Google Docs`.	4 / 5
	Total	15 / 20 Passed

Description

88%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: third-person, concrete, and explicitly paired with a 'Use when...' trigger clause that enumerates the natural user requests for each service. The only weaknesses are a few missing synonyms/file extensions in the trigger terms and slightly generic action phrases that create minor overlap risk with non-Google document skills.

Dimension	Reasoning	Score
Specificity	"Google Workspace CLI for Gmail, Calendar, Drive, Docs, Sheets, and Slides" names the domain, and "search email, schedule meetings, manage files, edit documents, update spreadsheets, or create presentations" lists six concrete actions with comprehensive coverage across all six services. Not 4 because coverage is broader than 'minor gaps' — every advertised service has a matching action.	5 / 5
Completeness	Explicitly answers 'what' ("Google Workspace CLI for Gmail, Calendar, Drive, Docs, Sheets, and Slides using gogcli") and 'when' ("Use when asked to search email, schedule meetings, manage files, edit documents, update spreadsheets, or create presentations") with concrete trigger phrases. Not 4 because the 'when' clause is already explicit and specific rather than merely present.	5 / 5
Trigger Term Quality	"search email", "schedule meetings", "edit documents" and the service names (Gmail, Sheets, Slides) are natural user phrases, but common variations are missing — no file extensions (.xlsx, .pptx, .docx) and no synonyms like "inbox", "deck", or "attachments". Not 5 because the natural-term coverage is good rather than comprehensive; not 3 because far more than 'some relevant keywords' are present.	4 / 5
Distinctiveness Conflict Risk	The named Google services give it a clear niche distinct from generic file or document skills, but generic action phrases like "manage files" and "edit documents" can fire for local-file or Word-style requests where the user never says "Google". Not 5 because that overlap with closely related document/file skills is more than minimal; not 3 because the service names anchor it far better than 'somewhat specific'.	4 / 5
	Total	18 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: odyssey4me/agent-skills
Path: skills/google/SKILL.md
Commit: 666332c

Reviewed: 4 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.