CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

google-docs

Prompt- and template-complete Google Docs creation and editing with explicit-instruction-authoritative structural preservation, including semantic roles, relationships, comparison dimensions, and instructed extensions; full-topology native-copy routing; source-grounded per-tab adaptation for past/example references; style-preserving hyperlink and table edits; canonical smart-chip-first authoring for dates and relevant supported people or Google resources; a file-backed advisory trusted read before existing-document writes; automatic protected-control awareness; direct connector APIs by default; DOCX-first import only when no supplied Google Doc template/reference constrains the output; and checked-in code mode only for exact native dropdown mutation. Use when Codex must create, edit, fill, adapt, redesign, or verify Google Docs without overriding explicit user/template instructions, adding unrequested document scope, or carrying stale reference facts into a new deliverable.

68

Quality

85%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

77%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 an unusually rigorous operational contract: routes are enumerated, workflows are sequenced with validation and re-read loops, and every reference path resolves to a real one-level-deep file. Its weakness is token efficiency — dense run-on sentences and the same preservation/stale-facts policies restated across multiple sections could be consolidated without losing any requirement.

Suggestions

Consolidate the stale-reference-facts policy (currently repeated across Default Routing, Native Reference Signature, Coverage-First Authoring, and Universal Completion Checks) into one stated invariant with cross-references.

Break up multi-clause sentences like the structural-preservation paragraph into short imperative bullets so each rule is individually scannable.

Inline one minimal `batchUpdate` request example for the most common edit so the core write primitive is executable without opening a reference file.

DimensionReasoningScore

Conciseness

All content is skill-specific policy rather than concepts Claude already knows, but it is noticeably over-dense: very long compound sentences (e.g., the multi-clause structural-preservation paragraph), and the stale-reference-facts rule is restated in at least four separate sections ('do not carry stale reference facts', 'never carry a reference-only schedule time...', 'no reference-only facts... remain'). Matches 'mostly efficient but... could be tightened' — not 2 because there is no filler or generic explanation, and not 4 because the repetition and run-on phrasing would trim substantially.

3 / 5

Actionability

Concrete, named primitives appear throughout — `mcp__codex_apps__google_drive._create_file`, `insertDate`/`insertPerson`/`insertRichLink`, `scripts/canonicalize_google_workspace_url.mjs`, `write_control.requiredRevisionId`, `getDocumentDropdowns`/`updateDocumentDropdown` — plus a routing table and a task-to-reference map. Mostly executable guidance with minor gaps — anchor 4, not 5, because full request-shape examples are deferred to reference files rather than shown inline, and the `host/docs-trusted-read-file-bridge.mjs` invocation details live entirely in the reference.

4 / 5

Workflow Clarity

Multi-step processes are explicitly sequenced (Direct-Request Workflow steps 1–11, Native Reference Signature steps 1–6) with explicit validation checkpoints and feedback loops: 'Re-read after substantive or index-shifting writes and continue from live indexes', fail-fast via `requiredRevisionId`, 'stop before destructive work and report the limitation', and a ten-item Universal Completion Checks checklist — matching the anchor-5 pattern.

5 / 5

Progressive Disclosure

The body is a routing overview with two dedicated navigation aids (Required Read Order keyed by route, and a Task Reference Map); all 18 cited `references/*.md` files exist and are exactly one level deep, and the `host/` bridge path is a real plugin file documented in `reference-trusted-read-wrapper.md`. 'Do not bulk-read the reference folder' and route-conditional reads are exemplary progressive disclosure — anchor 5.

5 / 5

Total

17

/

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.

The description is complete and highly specific, with an explicit 'Use when' trigger clause in third-person voice and clearly named capabilities. Its main weakness is extreme density — heavy compound coinages ('full-topology native-copy routing', 'explicit-instruction-authoritative') inflate length without adding trigger or distinctiveness value, and a few natural synonyms are missing.

Suggestions

Replace jargon compounds like 'explicit-instruction-authoritative structural preservation' and 'full-topology native-copy routing' with plain capability phrasing so the description reads cleanly at trigger-selection time.

Add one or two natural user synonyms (e.g., 'Google Doc', 'shared Drive document') to broaden trigger matching without adding length.

DimensionReasoningScore

Specificity

The description enumerates many concrete capabilities: 'Google Docs creation and editing', 'style-preserving hyperlink and table edits', 'smart-chip-first authoring for dates and... people', 'DOCX-first import', and 'code mode only for exact native dropdown mutation' — comprehensive concrete-action coverage matching the anchor-5 example.

5 / 5

Completeness

Both halves are explicit: a long 'what' listing the skill's capabilities and an explicit trigger clause 'Use when Codex must create, edit, fill, adapt, redesign, or verify Google Docs...' with concrete trigger phrases. This matches anchor 5; a 4 would require the 'when' to be weaker or less specific than it is.

5 / 5

Trigger Term Quality

Natural trigger terms are present ('Google Docs', 'create, edit, fill, adapt, redesign, or verify', 'template', 'dropdown') and a user would plausibly say them, but common variations like 'document', 'Google Drive', or file-form mentions are absent, and dense coinages like 'explicit-instruction-authoritative' add no trigger value. Good coverage with a few natural terms missing — anchor 4, not 5; not 3 since the core vocabulary users actually use is well represented.

4 / 5

Distinctiveness Conflict Risk

'Google Docs' names a clear niche with distinct triggers, but the description also covers DOCX import, template preservation, and generic document-editing language that overlaps with DOCX/Word or general document skills. Mostly distinct with minor overlap risk against closely related skills — anchor 4 rather than 5.

4 / 5

Total

18

/

20

Passed

Validation

93%

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

CriteriaDescriptionResult

relative_links

Relative link issues: 1 suspicious

Warning

Total

15

/

16

Passed

Repository

openai/plugins

Path

plugins/google-drive/skills/google-docs/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.