Token导航 LogoToken导航TokenDH.com
开发敏感数据clawhub未标认证来源可访问clear审计通过

lesecurelocallesecurelocal 开发

Agent Skill

lesecurelocal 用于补充开发相关能力,适合在 OpenClaw 中需要让 Agent 承接开发相关任务时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

3,370

周安装

135

GitHub Stars

公开资料未说明

下载量

1,091
OpenClaw

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

MIT-0

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:lesecurelocal(lesecurelocal 开发)
来源仓库:https://github.com/spalgorithm/lesecurelocal
安装命令:
openclaw skills install lesecurelocal
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 OpenClaw 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

ClawHubOpenClaw
openclaw skills install lesecurelocal

简介

lesecurelocal 支持本地或私有化部署环境下的文件与文件夹加解密。

  • 适用于企业内部系统间安全通信或离线数据存储保护需求。
  • 通过 clawhub 安装并使用 openclaw skills install lesecurelocal 命令部署。
  • 应确保终端设备符合安全基线要求,防止恶意软件窃取密钥材料。
  • 建议启用地理位置锁定等增强防护措施以提升整体安全性。

SKILL.md

name
LESecureLocal
description
LESecure Local/On-Prem — encrypt and decrypt data, files, and folders using the LE desktop tool with layered locks (pin, password, MFA, time lock, geo-location).
triggers

LESecureLocal — LESecure Local / On-Prem Skills

Encrypt and decrypt plain text, files, and folders using the LE desktop binary. No API key is needed — everything runs locally. No data leaves your machine.

Project Links

ResourceURL
Source code & documentation<https://github.com/SPAlgorithm/LE>
Cloud alternativeLESecureCl skill (uses the LESecure API)

ROUTING RULES (MANDATORY)

  • Files and folders MUST always use LESecure Local. Never use LESecure Cloud for file/folder encryption. If the user asks to encrypt files/folders via cloud, inform them: "File/folder encryption is only supported via LESecure Local (desktop)." and use this skill.
  • Safe-by-default flags for files/folders. Always pass -z (force overwrite) and — when the target is a directory — -n (recursive). Never pass -c (clean/delete source) or -j (trio = -z -c -n) without explicit user confirmation, because both delete the original file after encrypting. See the "Destructive Flags" rules below.
  • For plain text, ask the user: "Would you like to use LESecure Cloud (API) or LESecure Local (desktop)?" and proceed accordingly.
  • Current location queries — when the user asks "what's my current location", "whereami", "where am I", or any equivalent, run LE -7 and share the output. No other flags are needed. Note: LE -7 accesses device GPS — this is a privacy-sensitive operation. On first use in a session, inform the user: "This will query your device's GPS location via LE." Proceed only after acknowledgment.

Destructive Flags — -c and -j (MANDATORY)

  • -c (clean) deletes the source file after encryption or decryption. It is irreversible in-place data loss.
  • -j is a trio that includes -c, so it is also destructive.
  • Never silently add -c or -j. Before using either, ask the user explicitly, e.g.: "This will delete the source <file> after the operation. Confirm with 'yes, delete source' to proceed."
  • If the user does not confirm, use only -z (and -n for folders). The source stays on disk.
  • When the user explicitly asks for -j or "clean/delete source after", use -j and state in the response that the source was removed.

Binary Location (configuration)

The skill looks for the LE binary in this order:

  1. The LE_BIN environment variable, if set (e.g., export LE_BIN=/opt/le/LE).
  2. LE on PATH (via command -v LE).
  3. A user-supplied path if neither of the above resolves. In that case, ask the user for the binary path — do not guess or hardcode.

In examples below, LE is used as a shorthand for whichever path resolves. When actually invoking, expand it to the full resolved path so the command is reproducible.

# Resolve once, then reuse
LE_BIN="${LE_BIN:-$(command -v LE)}"
"$LE_BIN" --help

Date & Time Rules (MANDATORY)

All date/time handling for this skill follows these rules — no exceptions:

  1. Always use EST/EDT (America/New_York) to calculate and send dates. The LE tool interprets -l and -r in EST/EDT.
  2. Start time (-l) = current EST + 2 minutes by default. This buffer prevents the "date must be in future" error.
  3. End time (-r) = start time + the user's requested duration.
  4. Cross-platform time computation. Prefer Python because date flag syntax differs between BSD (macOS) and GNU (Linux). Python 3 is available on both.

Input safety: The <N> duration value is passed as sys.argv[1] and cast via int() inside the Python script — any non-integer input raises ValueError and the script exits without executing. Never concatenate or interpolate user input directly into the python3 -c string. Always pass values as positional arguments (sys.argv).

   # Start time (now + 2 minutes, EDT/EST) — no user input needed
   python3 -c "from datetime import datetime,timedelta; from zoneinfo import ZoneInfo; print((datetime.now(ZoneInfo('America/New_York'))+timedelta(minutes=2)).strftime('%Y/%m/%d %H:%M'))"

   # End time (now + 2 min + N minutes) — N is passed as argv[1], cast to int()
   python3 -c "import sys; from datetime import datetime,timedelta; from zoneinfo import ZoneInfo; N=int(sys.argv[1]); print((datetime.now(ZoneInfo('America/New_York'))+timedelta(minutes=2+N)).strftime('%Y/%m/%d %H:%M'))" <N>

   # End time (now + 2 min + N hours) — N is passed as argv[1], cast to int()
   python3 -c "import sys; from datetime import datetime,timedelta; from zoneinfo import ZoneInfo; N=int(sys.argv[1]); print((datetime.now(ZoneInfo('America/New_York'))+timedelta(minutes=2,hours=N)).strftime('%Y/%m/%d %H:%M'))" <N>

Fallback (date) — only if Python is unavailable: - macOS/BSD: TZ=America/New_York date -v+2M "+%Y/%m/%d %H:%M" - Linux/GNU: TZ=America/New_York date -d '+2 minutes' "+%Y/%m/%d %H:%M"

  1. Always display the window back to the user in EDT/EST.

Two Modes

1. PlainText Mode (--PlainText / -p)

Encrypt/decrypt inline strings. The LE binary expects the data wrapped in triple single quotes ('''...''').

Input Sanitization (MANDATORY)

Never interpolate raw user input directly into the shell command. The '''...''' quoting breaks if the data contains single quotes, enabling shell injection. Before building the command:

  1. Validate: reject or escape any single quotes (') in the user's plaintext. Replace each ' with '\'' (end quote, escaped literal quote, reopen quote).
  2. Alternatively, use a shell variable to isolate user data from the command string:
   # Store user data in a variable — shell expansion is safe inside triple quotes
   LEDATA='user provided text here'
   LE -e "'''${LEDATA}'''" <LOCK_FLAGS> --PlainText
  1. Never use eval or backtick interpolation with user-supplied text.
# Encrypt (with sanitized data)
LE -e '''<SANITIZED_DATA>''' <LOCK_FLAGS> --PlainText

# Decrypt (encrypted output is safe — no special chars)
LE -d '''<ENCRYPTED_DATA>''' <LOCK_FLAGS> --PlainText

2. File / Folder Mode

Default flags (safe):

  • File: -z
  • Folder: -z -n

Destructive extras (only with explicit user confirmation, see rules above): add -c to also delete the source, or use -j (= -z -c -n).

# Safe file encrypt / decrypt (source file preserved)
LE -e <FILE>   <LOCK_FLAGS> -z
LE -d <FILE.letxt> <LOCK_FLAGS> -z

# Safe folder encrypt / decrypt (source folder preserved)
LE -e <FOLDER> <LOCK_FLAGS> -z -n
LE -d <FOLDER> <LOCK_FLAGS> -z -n

# Destructive — only after explicit user confirmation
LE -e <FILE_OR_FOLDER> <LOCK_FLAGS> -j

Naming notes:

  • Encrypted files get a .le prefix on the extension (e.g., example.txt becomes example.letxt); use the .letxt filename when decrypting.
  • For folders, the individual files inside get the .le prefix on their extensions. The folder name itself stays the same.

Available Locks

FlagLock TypeValueExample
-1Pin/CodeNumeric string"1122"
-wPasswordPassword file (.letxt) or passphrasepass.letxt
-2MFAPhone number (E.164)"+19199870623"
-3OTPOTP code for decryption"123456"
-lTime lock startYYYY/MM/DD HH:MM"2026/04/12 17:41"
-rTime lock endYYYY/MM/DD HH:MM"2027/04/12 17:36"
-bLocation lock — use existing .lecsv key file (encrypt only; omit on decrypt)Path to .lecsv filelocation.lecsv
-vLocation lock — create a new .lecsv key file from a GPS CSV (switch, no value)(no value)-v

Additional Flags

FlagPurposeSafety
-zForce — overwrite existing encrypted fileSafe
-nRecursive — process folders recursivelySafe
-cClean — delete source after encrypt/decryptDESTRUCTIVE — opt-in with confirmation
-jTrio = -z -c -nincludes delete-sourceDESTRUCTIVE — opt-in with confirmation
-iGet info on an encrypted fileSafe (read only)
-oSpecify output file nameSafe
-7Print the device's current GPS location (no other flags needed)PRIVACY-SENSITIVE — requires user consent on first use

MFA Workflow

  1. Encrypt with MFA: Use -2 "+1XXXXXXXXXX" to register the phone number.
  2. Decrypt with MFA: First run decrypt with -4 <encrypted_file> to trigger OTP delivery, then run again with -3 <OTP_CODE>.

Examples

PlainText — Pin only

LE -e '''hello world''' -1 "1234" --PlainText
LE -d '''<ENCRYPTED>''' -1 "1234" --PlainText

PlainText — All locks

LE -e '''secret data''' -w pass.letxt -1 "1122" -2 "+19199870623" -l "2026/04/12 17:41" -r "2027/04/12 17:36" --PlainText

File — Pin only (safe, source preserved)

LE -e /path/to/myfile.txt -1 "1234" -z
LE -d /path/to/myfile.letxt -1 "1234" -z

Folder — Pin + Password (safe, sources preserved)

LE -e /path/to/my_folder -w pass.letxt -1 "1234" -z -n
LE -d /path/to/my_folder -w pass.letxt -1 "1234" -z -n

File — destructive (user asked to delete source)

# Only after explicit user confirmation
LE -e /path/to/myfile.txt -1 "1234" -j

Get info on encrypted file

LE -i /path/to/myfile.letxt

Get current device location

Requires user consent on first use in a session (privacy-sensitive — accesses device GPS).

LE -7

Workflow

  1. Determine the mode: PlainText (--PlainText) for inline strings, or File/Folder for files and directories.
  2. Resolve the binary via $LE_BIN, command -v LE, or ask the user.
  3. Gather lock inputs: Which locks to apply and their values.
  4. Pick safe defaults: -z for files; -z -n for folders. Do not add -c or -j unless the user explicitly confirmed source deletion.
  5. Build the command with the appropriate flags.
  6. Execute via Bash and return the result.
  7. For decryption, remind the user they need the same lock values used during encryption.

Important Notes

  • No API key is needed — LE runs entirely locally.
  • Phone numbers for MFA (-2) must be in E.164 format.
  • Time lock dates use YYYY/MM/DD HH:MM format. Follow the Date & Time Rules above.
  • Time locks require both -l (start) and -r (end).
  • The password file (.letxt) should be an encrypted password file created with LE -e pass.txt -q.
  • Geo-location locks work in two stages: create a key file once, then reuse it to lock as many files/folders as you want.

Stage 1 — Create the .lecsv key file from a GPS CSV (-v): - Input: a plain CSV of GPS locations with distance (e.g., location.csv). - -v is a switch (no value); LE produces location.lecsv alongside the input. - MUST be paired with -1 (pin) or -2 (MFA) — otherwise LE errors with "Either Pin or MFA should be enabled for Password/Location file".

  LE -e location.csv -v -1 1122 -z
  LE -e location.csv -v -2 "+1YourPhoneNumber" -z

Stage 2 — Use the .lecsv key file to lock files/folders (-b): - -b <path.lecsv> is used only on encryption. - On decryption, do NOT pass -b — LE reads the embedded location reference from the encrypted file itself. Just run LE -d <file> -z. - No pin/MFA pairing required — the key file is self-contained.

  # Encrypt (pass -b with the key file)
  LE -e example.txt -b location.lecsv -z

  # Decrypt (do NOT pass -b)
  LE -d example.letxt -z

适合场景

01

OpenClaw 用户查找和安装 Skill 时

02

用户想查找某类 Agent Skill 时

03

需要根据任务场景推荐可安装能力包时

04

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

补充不同宿主或平台的使用分布数据

能力 5

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

OpenClaw

98.41%
按下载量换算1,074

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

通过

权限和风险

敏感数据

该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills