CtrlK
BlogDocsLog inGet started
Tessl Logo

banana-cli

CLI tool for creating, managing, and exporting AI-generated presentations via the Banana Slides API. Use when the user asks to: (1) generate a PPT/presentation/slides from an idea, outline, or description, (2) export a project to PPTX, PDF, or images, (3) batch-generate multiple presentations, (4) manage projects, pages, materials, or templates programmatically, (5) renovate/redesign an existing PPT or PDF, (6) edit slide images with natural language.

74

Quality

93%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

87%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 efficient, highly actionable body with executable commands throughout, a proper validation gate on the environment check, and exemplary progressive disclosure of setup details into references/setup.md. The main gap is the batch-generation workflow, which lacks any post-run verification or failure-recovery step, capping workflow clarity at 3.

Suggestions

Add a validation step after batch generation, e.g. 'After run jobs completes, check report.json for failed jobs (jq '.jobs[] | select(.status != "success")') and re-run failed entries — run jobs skips already-succeeded jobs via --state-file'.

State what a failed export or generation looks like (error exit code, JSON error shape) and what to do about it, so the end-to-end workflow has a recovery loop for its async tasks.

Clarify in the batch section whether --state-file makes re-runs idempotent, so an agent knows it can safely retry the whole file.

DimensionReasoningScore

Conciseness

The body is lean and every token earns its place: all sections are runnable commands or operational facts Claude cannot know ('accept short prefixes (like git short hashes)', "--pages is a hint to the AI", 'Progress lines go to stderr (format: [PROGRESS] ...)', 'Config priority: CLI args > env vars > TOML'). No concept explanations or padding, so it does not fall to 4.

5 / 5

Actionability

Everything is copy-paste executable: a complete end-to-end workflow (create project via --json + jq, set working project, run full workflow, export), batch generation with a concrete jobs.jsonl heredoc, renovation command, and JSON piping examples. Specific examples cover the common cases, matching the top anchor.

5 / 5

Workflow Clarity

The main workflow is clearly sequenced and gated by an explicit health check ('Do not proceed until the health check passes'), but the batch-generation workflow ('banana-cli run jobs --file jobs.jsonl --report report.json') is a batch operation with no validation/verification step — nothing instructs checking report.json for failed jobs or retrying them. Per the rubric, missing validation in batch operations caps workflow clarity at 3.

3 / 5

Progressive Disclosure

The body is a concise overview with a single clearly-signaled, one-level-deep reference ('Read [references/setup.md](references/setup.md)') that exists as a real bundle file containing the offloaded setup steps; there is no inlined content that belongs in a separate file and no nested references. This matches the actual bundle structure, not just the reference links.

5 / 5

Total

18

/

20

Passed

Description

96%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: concrete capabilities, explicit third-person 'what', and a well-enumerated 'Use when' trigger list covering the six main use cases with synonyms and format terms. The only residual risk is minor overlap with generic presentation-file skills.

DimensionReasoningScore

Specificity

The description lists multiple concrete actions — 'creating, managing, and exporting AI-generated presentations', 'export a project to PPTX, PDF, or images', 'batch-generate', 'renovate/redesign an existing PPT or PDF', 'edit slide images with natural language' — with comprehensive coverage and third-person voice throughout. It does not fall to 4 because there are no minor gaps in action coverage.

5 / 5

Completeness

It explicitly answers both 'what' ('CLI tool for creating, managing, and exporting AI-generated presentations via the Banana Slides API') and 'when' via the enumerated 'Use when the user asks to: (1)...(6)' trigger list with concrete phrases. Anchor 4 would require the 'when' to be less explicit.

5 / 5

Trigger Term Quality

Natural terms users would say are comprehensively covered with synonyms and format extensions: 'PPT', 'presentation', 'slides', 'PPTX', 'PDF', 'images', 'outline', 'redesign'. Not 4 because the common variations (including file-format names) are present rather than 'a few missing'.

5 / 5

Distinctiveness Conflict Risk

The 'Banana Slides API' anchor and programmatic framing give it a mostly distinct niche, but triggers like 'generate a PPT/presentation' and 'export a project to PPTX' carry minor overlap risk with a generic .pptx document-manipulation skill. Not 5 because that closely related skill family exists; not 3 because the AI-generation/renovation framing and API reference clearly differentiate it.

4 / 5

Total

19

/

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
Anionex/banana-slides
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.