#secrets-management

A simple MacOS helper for safer exposure of secrets to your agents

🕶️ Secrets your AI agent can USE but never SEE — Claude Code skill + zero-dependency macOS Keychain CLI. Env injection, pointer manifest, prompt-injection tripwire.

Runtime infrastructure for AI agents — lifecycle, supervision, secrets, and deterministic time-travel replay. Built in Rust.

Self-host secrets manager with an AI-safety layer. Aliases instead of values; AI agents never see real credentials. Cloud option coming.

Trusty Squire signs up / in to websites for you so you don’t have to.

Save a secret: a small window opens that Claude cannot see, you paste there, it lands in 1Password. Read it back: any item in your vault loads into Claude with one fingerprint check, names only, never the values. Keys you paste in the chat are deleted before Claude reads them. Mac, Windows, Linux. MIT.

Agent Skill for using secrets from a local KeePassXC database — kp:// references, in-process resolution, env injection and output redaction. Works in Claude Code, Codex, Cursor, Gemini CLI and any AGENTS.md harness

Lets Claude Code read service API keys/tokens from a running, unlocked KeePass 2 instance via the KeePassNatMsg plugin

Starter for AI agents that never hold their own secrets: screen untrusted input and fetch API keys just-in-time with 1Claw. Part 1 runs with no account.

🔐⛓️ Local-first credential vault for AI agents — encrypted storage, scoped tokens, hash-chained audit. A copied URL is a reference, never a credential.

Deterministic 3-layer secrets protection for Claude Code: permissions.deny + PreToolUse hook (Read/Edit/Grep/Bash/PowerShell) + CLAUDE.md redaction policy. MCP-safe by design. Born when .claudeignore turned out to be unenforced and prompt-only rules kept leaking secrets into AI context.

多 Agent 团队的本机密钥库：密钥加密 + 框架模型一键切换（Codex / Claude）。致用，不吹爆款。