← 返回 Skills 市场

Tene CLI — Local-First Secrets

Name: Tene CLI — Local-First Secrets
Author: tomo-kay

作者 agent-kay · GitHub ↗ · v1.0.3 · MIT-0

macoslinux ✓ 安全检测通过

总下载

当前安装

版本数

在 OpenClaw 中安装

/install tene-cli

功能描述

Local-first encrypted secret management with the tene CLI. Activate when the user mentions secrets, API keys, credentials, tokens, .env files, environment va...

安全使用建议

This skill appears to be what it says: documentation-only guidance for using the tene CLI. Before installing: review the installer script (https://tene.sh/install.sh) rather than piping it blindly into sh; prefer building from source or verifying a release checksum if you want higher assurance. Note that tests/docs mention an Anthropic API key for evaluation — that is not required to use tene itself. Follow the skill's safety rules (never paste secrets into chat, never run `tene get` in the agent context, use `tene run --` to inject secrets) and be deliberate about whether your team wants the encrypted `.tene/vault.db` committed into source control for CI workflows.

功能分析

Type: OpenClaw Skill Name: tene-cli Version: 1.0.3 The tene-cli skill bundle provides a secure interface for an AI agent to manage secrets using the 'tene' CLI tool. It features extensive safety guardrails in SKILL.md and tests/test.md that explicitly prevent the agent from outputting plaintext secrets or accessing the encrypted vault directly. The installation process via https://tene.sh/install.sh is a standard practice for CLI tools, and the overall logic is focused on enhancing security rather than compromising it.

能力标签

requires-walletrequires-sensitive-credentials

能力评估

✓ Purpose & Capability

Name, description, and required binary (tene) are coherent. The skill documents exactly the commands a user would need for local secret management and does not request unrelated OS credentials, binaries, or config paths.

ℹ Instruction Scope

SKILL.md is prescriptive and scoped to secret-management tasks (init, set, list, run, import/export with encrypted flag, CI patterns). Example flows include reading .env files for import, deleting those plaintext files, and wrapping commands with `tene run --` — all consistent with the stated purpose. The tests mention using an external LLM backend (Anthropic) requiring ANTHROPIC_API_KEY for behavioral evaluation; that is only part of the test harness and not a runtime requirement of the skill, but it is surprising and worth noting.

⚠ Install Mechanism

The skill recommends installing via the project's install script (curl -sSfL https://tene.sh/install.sh | sh). This is a pragmatic, common installer for CLI tools but is higher-risk than a vetted package (e.g., Homebrew or GitHub release checksum) because it executes a remote script. The SKILL.md also documents building from source (go install) which is a safer alternative.

ℹ Credentials

The skill itself requests no environment variables or credentials, which is proportional. Example/CI docs rightly instruct storing a TENE_MASTER_PASSWORD secret in CI. Tests reference ANTHROPIC_API_KEY (and an EVAL_BACKEND) for behavioral tests — these are test harness needs and not required to use the skill; mention of them could confuse users into thinking the skill needs an LLM API key.

✓ Persistence & Privilege

always:false and normal autonomous invocation are used. The skill does not request permanent system-wide privileges or attempt to modify other skills' configs. It does recommend committing the encrypted vault in some workflows (explicitly discussed and justified).

如何使用

确保已安装 OpenClaw（本地或 Docker 部署）
在对话框中输入安装命令：/install tene-cli
安装完成后，直接呼叫该 Skill 的名称或使用 /tene-cli 触发
根据 Skill 的参数说明提供必要输入，即可获得结构化输出

版本历史

v1.0.3

tene-cli 1.0.3 Changelog - Updated SKILL.md with detailed usage instructions and strict AI safety rules. - Clarified activation triggers and restrictions for when the skill should be used. - Added comprehensive documentation for all main tene commands and workflows (init, set, list, run, import, export, env, passwd, etc.). - Reiterated critical safety practices, including never leaking plain secrets and only injecting via `tene run --`. - Provided updated install and migration guidance. - 100% coverage of active tene CLI commands (init, set, get, list, delete, run, import, export, env, passwd, recover, version, update, whoami) - 5 baked-in AI safety rules: never run `tene get`, never plain `tene export`, never read `.tene/`, never pass secrets as CLI args, use `tene list` for introspection - 3 end-to-end examples (init, .env migration, multi-env CI/CD) and 6 expected-behavior test cases - Verified: 19/19 offline assertion self-tests + 6/6 live behavioral eval (3 consecutive runs, 100%) via `claude -p`

元数据

Slug tene-cli

版本 1.0.3

许可证 MIT-0

累计安装 0

当前安装数 0

历史版本数 1

常见问题

Tene CLI — Local-First Secrets 是什么？

Local-first encrypted secret management with the tene CLI. Activate when the user mentions secrets, API keys, credentials, tokens, .env files, environment va... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 88 次。

如何安装 Tene CLI — Local-First Secrets？

在 OpenClaw 或 Claude Code 对话框中运行命令「/install tene-cli」即可一键安装，无需额外配置。

Tene CLI — Local-First Secrets 是免费的吗？

是的，Tene CLI — Local-First Secrets 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。

Tene CLI — Local-First Secrets 支持哪些平台？

Tene CLI — Local-First Secrets 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（macos, linux）。

谁开发了 Tene CLI — Local-First Secrets？

由 agent-kay（@tomo-kay）开发并维护，当前版本 v1.0.3。