CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

deliverables

How to write a document the user will read (report, itinerary, comparison, plan, notes) and save it to their Google Drive. Use when a thread's result is a document rather than a code change.

Quality

84%

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

86%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.

An exceptionally lean, instruction-only skill that gives fully specific rules for naming, formatting, content structure, Drive saving, and reply etiquette with no wasted tokens. The only meaningful gap is the absence of a demonstrated tool call or output example, which keeps actionability and workflow clarity at 4 rather than 5.

DimensionReasoningScore

Conciseness

Every bullet is a dense directive or constraint Claude could not infer on its own — "named after the subject (`alaska-itinerary.md`, never `report.md`)", "stay under 14 KB", "Markdown → Doc, CSV → Sheet, `.pptx` → Slides" — with zero concept explanation or padding. The one justification offered ("Because the user sees the Drive link next to your reply") is necessary context for the cover-note rule, not filler, so the body clearly matches the lean/efficient anchor rather than the merely-efficient one.

5 / 5

Actionability

Guidance is concrete and executable: specific file-naming examples, a hard 14 KB cap, exact tool names (`save_to_drive`, `present_card`), parameter specs ("with `path` (relative to your working directory) and the `#` title as `title`"), and a format mapping. It falls short of the fully-executable anchor only because no example tool invocation or document skeleton is shown — the `save_to_drive` call and reply format are described but never demonstrated — while clearly exceeding the pseudocode/incomplete anchor.

4 / 5

Workflow Clarity

The bullets read as an ordered workflow (create file → lead with the answer → update in place → save when final → reply as cover note) with an explicit gating checkpoint ("call it once the deliverable is final") and a guard ("Never save scratch notes or intermediate files"). Minor validation gaps — no post-save verification and no explicit size check before upload — keep it below the explicit-validation-loop anchor, and the non-destructive single-file scope means the destructive/batch cap does not apply.

4 / 5

Progressive Disclosure

The skill is a 14-line body with five bolded, clearly labeled sections and no bundle files (references/, scripts/, assets/ are absent), so nothing needs offloading. Per the rubric's simple-skill guidance, an under-50-line skill with no need for external references scores 5 on well-organized sections alone, which this body has.

5 / 5

Total

18

/

20

Passed

Description

73%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 well-constructed description with an explicit, genuinely distinctive trigger condition and good natural keyword coverage. Its main weakness is that the capability list stops at two actions, underselling the skill's actual scope and leaving it slightly less concrete than it could be.

Suggestions

Enumerate the skill's concrete behaviors more fully — e.g., "draft, name, and iteratively update a Markdown deliverable, then upload it to Google Drive as a Doc, Sheet, or Slides file" — to raise specificity beyond two generic actions.

Add concrete user-mention trigger phrases to the 'Use when' clause, such as "when the user asks for a report, itinerary, plan, or comparison", so the when-condition matches how users actually phrase requests.

Include a couple of common synonyms for the deliverable types (e.g., "summary", "write-up", "memo") to broaden natural trigger coverage.

DimensionReasoningScore

Specificity

It names the domain and exactly two concrete actions — "write a document the user will read (report, itinerary, comparison, plan, notes)" and "save it to their Google Drive" — matching the "1-2 concrete actions, not comprehensive" anchor. The enumerated document types add flavor, but the description omits the skill's other behaviors (iterative updating, format conversion to Doc/Sheet/Slides, cover-note replies), so it does not reach the "several specific actions" anchor.

3 / 5

Completeness

Both halves are explicit: the what ("write a document the user will read ... and save it to their Google Drive") and the when ("Use when a thread's result is a document rather than a code change"). It stops short of the top anchor because the when-clause states a condition rather than concrete user-mention trigger phrases (e.g., "when the user asks for a report or itinerary"), and it is well above the weak/implicit-when anchor.

4 / 5

Trigger Term Quality

Natural terms users would actually say are well covered: "report, itinerary, comparison, plan, notes", "document", and "Google Drive". Common synonyms and variants ("summary", "write-up", "memo", "brief", "deck") are missing, which fits the good-coverage-with-a-few-gaps anchor rather than comprehensive synonym/extension coverage.

4 / 5

Distinctiveness Conflict Risk

The trigger "Use when a thread's result is a document rather than a code change" is a highly discriminative condition that explicitly carves the skill apart from code-change skills, and the Drive-upload deliverable framing gives it a clear niche with minimal overlap risk against generic document or Drive-file skills.

5 / 5

Total

16

/

20

Passed

Validation

100%

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

google/adk-samples

Path

core/python/adk-voice-agent/sandbox/skills/deliverables/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.