Skill Packager

Perplexity Computer · agentskills.io · ZIP Export

◐ 5/9 checks

🏷

Skill Identity

REQUIRED

Skill NameLowercase, hyphens only, ≤64 chars

Risk Tier

L0=safe · L1=low · L2=elevated · L3=destructive

Description — routing trigger0/50 words

Start with "Load when" or "Use when" · ≤50 words · this IS the routing signal

Author

Version

License

📝

SKILL.md Body

REQUIRED

Constraints · Workflow · Gotchas. Drag a .md file, click Upload, or paste directly. Use Add More to append additional instruction blocks.

Instruction Body

⤓ Drag a file here or click to upload

📚

Reference Files

OPTIONAL

Brand guidelines, legal docs, schemas — loaded on demand. Packaged into references/

Reference

⤓ Drag a file here or click to upload

🗂

Asset Files

OPTIONAL

Templates, logos, icons — binary or text files the agent copies/modifies. Packaged into assets/

Asset

⤓ Drag a file here or click to upload

⚙️

config.json

OPTIONAL

First-run setup params. Stored at zip root. Not injected into model context. Drag a .json file, upload it, or paste JSON directly.

config.json

⤓ Drag a file here or click to upload

🌌

Perplexity Space Metadata

AUTO-GENERATED

Ready-to-paste details for creating a new Perplexity Space configured for this skill.

Space Title

Space Description

Custom Instructions

Pre-flight

5/9

✗

Name: lowercase-hyphenated, 2–64 chars

(empty)

✓

Name: no consecutive hyphens

✗

Description starts with "Load when" or "Use when"

✗

Description ≤1024 chars (0)

✓

Description ≤50 words (0 used)

✗

Instruction body is not empty

✓

config.json slots are valid JSON

✓

Est. size ≤10 MB (0 B)

✓

Total files ≤100 (1)

ZIP Structure

📄SKILL.mdroot ✓

Add refs / assets to see full tree

Body Blocks

References

Assets

Config files

Upload to Perplexity

Computer → Skills → Create skill → Upload a skill

SKILL.md at ZIP root · ≤10 MB · ≤100 files