secrets · ulpi-io — Claude Code skill: what it does, how to install
A skill that drives the secrets CLI: add, rotate, list and toggle local service credentials while values stay out of commands and chat.
no license 148 installs ★ 5 updated 2026-08-31
npx skills add ulpi-io/skills --skill secrets -a claude-codeWhat secrets does
The skill gives an agent a fixed routine for credential work. You name the action, it picks the safe command shape, and the value goes in through the CLI's interactive prompt instead of a command line argument.
Covered actions are add, remove, rotate, list, masked status, enable, disable and project init. Only two tools are allowed: Bash and Read. A vague request gets a question back before anything runs, so the wrong service doesn't get touched.
Switching a service on does not prove the integration works, and the skill reports it that way: status and enablement, nothing more. It also stays dormant until you call it, because model invocation is disabled in its frontmatter.
What it helps with
- store a GitHub token without leaving it in shell history
- rotate a key and keep the value out of chat output
- list which services are already configured
- switch an MCP integration on or off
- check vault status before an agent run
Example prompts for the agent
- Add credentials for github with
/secrets add github; I'll type the value into the prompt. - Which services are configured and enabled? Run
/secrets list. - Turn on the github integration for MCP tooling:
/secrets enable github.
How to install secrets
1Skills CLI (Vercel)
npx skills add ulpi-io/skills --skill secrets -a claude-code2Manually: put the files in place
~/.claude/skills/secrets/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/secrets/
1Skills CLI (Vercel)
npx skills add ulpi-io/skills --skill secrets -a codex2Manually: put the files in place
.agents/skills/secrets/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add ulpi-io/skills --skill secrets -a cursor2Manually: put the files in place
.cursor/skills/secrets/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install ulpi-io/skills secrets2Manually: put the files in place
.github/skills/secrets/SKILL.md1Skills CLI (Vercel)
npx skills add ulpi-io/skills --skill secrets -a gemini-cli2Platform CLI
gemini skills install https://github.com/ulpi-io/skills --path secretsManually: put the files in place
.devin/skills/secrets/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add ulpi-io/skills --skill secrets -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 secrets
How do I install secrets in Claude Code?
Run npx skills add ulpi-io/skills --skill secrets -a claude-code. The skill file lands at ~/.claude/skills/secrets/SKILL.md.
What does secrets do?
It walks an agent through credential operations against the local secrets CLI: add, remove, rotate, list, enable and disable. Only service names and status come back.
Can the skill show me a stored secret value?
No. --reveal is off limits, and values never enter arguments, logs or output. You get the service name and whether it is enabled.
Similar skills
All skills for Security →secrets by other authors comes first.