plugins/nopeek/skills/nopeek/SKILL.md
Secret-safe Claude Code sessions via nopeek CLI. Use when user asks about secrets, .env loading, credential safety, or keeping API keys out of conversation context.
npx skillsauth add spences10/claude-code-toolkit nopeekInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Secure proxy between Claude Code and your secrets. Claude knows key names, never key values.
Every Bash tool output is sent to Anthropic's API and retained 30+ days. If a CLI prints a token, that token is stored on Anthropic's servers. nopeek prevents this.
All nopeek commands are safe inside Claude Code — they never output secret values, only key names.
npx nopeek init
Detects installed cloud CLIs (aws, hcloud, kubectl, etc.), checks their auth config, and migrates to profile-based auth where needed.
npx nopeek load .env
npx nopeek load .env --only DATABASE_URL,API_KEY
Injects values into the session via CLAUDE_ENV_FILE (when available) or outputs a source command. Only key names appear in output — never values.
npx nopeek set MY_API_KEY --from-env
npx nopeek set STRIPE_KEY --value "sk_live_..."
Stored keys auto-load on every Claude Code session via the SessionStart hook.
npx nopeek status
npx nopeek list
| Hook | Event | Purpose |
|------|-------|---------|
| session-load.sh | SessionStart | Loads stored keys + CLI profiles into CLAUDE_ENV_FILE |
| redact-output.sh | PreToolUse(Bash) | Wraps cloud CLI output through secret pattern redaction |
CLAUDE_ENV_FILE so they never appear in output at all.|, >, or < to avoid breaking complex command semantics.npx nopeek audit
Scans current directory for .env files containing secrets and checks .gitignore coverage.
Keys not loading on session start:
npx nopeek status — are keys stored?/plugin list/reload-pluginsCloud CLI still showing secrets:
npx nopeek load .env to make secrets available as env vars, then use $VAR_NAME in commandsdevelopment
Design and create Claude Skills using progressive disclosure principles. Use when building new skills, planning skill architecture, or writing skill content.
testing
Verify sources before presenting findings. Use when asked to research links or documentation.
data-ai
Analyze session history for learnings and persist to skills. Solves "memory zero" - correct once, never again.
tools
Validate, test, and distribute Claude Code plugins and marketplaces. Use when developing plugins, debugging validation errors, or preparing for distribution.