my-venice-skill — Claude Code skill: a Venice API skill template
A starter skill file for agents: section layout, Venice API request samples and an error table. Useful when you write your own Venice skill.
npx skills add veniceai/skills --skill template -a claude-codeWhat my-venice-skill does
This is a template, not a working tool. The SKILL.md skeleton covers endpoints in a table, a quick start with curl and an SDK snippet, required and optional parameters, a sample response, an error table and a gotchas list. The frontmatter description tells the agent when to load the skill, so naming exact endpoints and scenarios matters for selection.
The quick start calls https://api.venice.ai/api/v1/example with an Authorization: Bearer $VENICE_API_KEY header. Real use therefore needs a Venice key in the VENICE_API_KEY environment variable. The error table maps 400 to bad input and 429 to a rate limit, where the advice is to back off with jitter.
Install runs through npx skills add veniceai/skills --skill template -a <agent>, and GitHub Copilot uses gh skill install instead. Paths differ per editor: Claude writes to ~/.claude/skills/template/SKILL.md, Cursor to .cursor/skills/template/SKILL.md, Codex to .agents/skills/template/SKILL.md. The file ends with cross-links to sibling skills such as venice-errors and venice-billing, which you will have to swap for your own.
What it helps with
- start a new Venice API skill without an empty file
- describe endpoints and parameters so the agent picks the skill reliably
- show the agent request samples that read the key from an environment variable
- map API error codes to causes and fixes
- link the new skill to neighbouring ones with cross-references
Example prompts for the agent
- Fill in
SKILL.mdfor a new skill: endpointPOST /example, required params and a sample response - Add a
429row to the errors section and explain backing off with jitter - Swap the cross-links at the bottom of the template for my own skills
How to install my-venice-skill
1Skills CLI (Vercel)
npx skills add veniceai/skills --skill template -a claude-code2Manually: put the files in place
~/.claude/skills/template/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/template/
1Skills CLI (Vercel)
npx skills add veniceai/skills --skill template -a codex2Manually: put the files in place
.agents/skills/template/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add veniceai/skills --skill template -a cursor2Manually: put the files in place
.cursor/skills/template/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install veniceai/skills template2Manually: put the files in place
.github/skills/template/SKILL.md1Skills CLI (Vercel)
npx skills add veniceai/skills --skill template -a gemini-cli2Platform CLI
gemini skills install https://github.com/veniceai/skills --path templateManually: put the files in place
.devin/skills/template/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add veniceai/skills --skill template -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about my-venice-skill
How do I install my-venice-skill in Claude Code?
Run npx skills add veniceai/skills --skill template -a claude-code. The file lands at ~/.claude/skills/template/SKILL.md.
What does my-venice-skill do?
It hands you a skill skeleton for the Venice API: an endpoint table, request samples, and breakdowns of parameters, responses and errors. You fill it with your own content.
What do the request samples need to run?
A Venice key in the VENICE_API_KEY environment variable, since the sample passes it in the Authorization: Bearer $VENICE_API_KEY header.
Source text
SKILL.md · MIT
---
name: my-venice-skill
description: One or two sentences describing exactly when an agent should load this skill and what it covers. Mention the specific endpoints, parameters, or scenarios so the agent can confidently pick it — vague descriptions hurt skill selection.
---
# My Venice skill
One-paragraph summary of what this skill teaches the agent. Focus on which Venice API surface it maps to and when to prefer it over related skills.
## Endpoints
| Method | Path | Notes |
|---|---|---|
| `POST` | `/example` | Required params, authentication, cost model. |
## Quick start
```bash
curl -X POST https://api.venice.ai/api/v1/example \
-H "Authorization: Bearer $VENICE_API_KEY" \
-H "Content-Type: application/json" \
-d '{ "foo": "bar" }'
```
```ts
// SDK or fetch example
```
## Parameters
### Required
- `field_a` — what it is, valid range, default.
### Optional
- `field_b` — …
## Response
```json
{
"example": "response"
}
```
## Errors
| Status | Cause | Fix |
|---|---|---|
| `400` | Bad input | Validate against the schema. |
| `429` | Rate limit | Back off with jitter. |
## Gotchas
- Specific edge cases, rate-limit caps, model constraints.
- Cross-links to related skills: [`venice-errors`](../venice-errors/SKILL.md), [`venice-billing`](../venice-billing/SKILL.md), etc.
© veniceai · MIT · veniceai/skills
