Agent Skills › simplexai-labs/linger › lev8-search

lev8-search

GitHub

通过 lev8 CLI 执行联系人查询、实体搜索、深度研究及数据提取,支持断点续传和设备登录认证。

skills/builtin/lev8-search/1.0.1/SKILL.md simplexai-labs/linger

Trigger Scenarios

需要进行深度信息检索或数据挖掘 请求获取特定实体或联系人的详细资料 需要恢复中断的 API 请求任务

Install

npx skills add simplexai-labs/linger --skill lev8-search -g -y
More Options

Non-standard path

npx skills add https://github.com/simplexai-labs/linger/tree/main/skills/builtin/lev8-search/1.0.1 -g -y

Use without installing

npx skills use simplexai-labs/linger@lev8-search

指定 Agent (Claude Code)

npx skills add simplexai-labs/linger --skill lev8-search -a claude-code -g -y

安装 repo 全部 skill

npx skills add simplexai-labs/linger --all -g -y

预览 repo 内 skill

npx skills add simplexai-labs/linger --list

SKILL.md

Frontmatter
{
    "name": "lev8-search",
    "description": "Use the lev8 CLI for contacts, entity searches, deep research, schema-based enrichment and social\/GitHub\/web extraction through Lev8 External API, recover interrupted requests with private receipts, and complete browser device login. Does not send emails or messages."
}

lev8 CLI

When to use

Use the examples directly for known commands; run lev8 <command> --help only for unfamiliar flags or a version mismatch. Prefer --json (stdout is one compact line) and read ok, error and meta; HTTP 200 or a task ID is not completed work.

Availability

Cloud Host includes lev8 0.5.0 and supplies only CI connection defaults, never credentials. Local Computer uses the user's own CLI and API configuration (CLI default: production). If missing or older than 0.5.0, ask the user to install/upgrade; do not install it yourself or silently switch APIs.

Credentials

  • If credentials are missing, run lev8 auth device start --name 'Linger' --storage config --json. Send data.verification_uri_complete and data.user_code to the user to compare, and keep data.authorization_id. data depends on status: pending has approval details and expires_at; authorized has source, masked token, completed_at and config_path (file storage); terminal states have only ID and status, with guidance in error. No links remain after the attempt ends.
  • Finish the turn while they approve, then run lev8 auth device complete --id <authorization_id> --json. It waits at most 30 seconds; exit 7 means run it again later with the same ID. Without --json, start prints the same essentials.
  • The approving account becomes this Context's credential and billing account. Ask the requesting user to approve with their own account and only the link they requested. Opening the link never approves by itself. Never ask for an API key or put tokens in arguments, objectives, result files or progress messages.
  • Other login paths, storage and config: auth-and-config.md.

Running commands

Do not query or display credit balances or inspect billing tools. Follow the authorized search scope and never fan out queries to beat limits. Starting and resuming are different actions.

lev8 deep-search --query 'research question' --json --stdout-limit 16384
  • Use --stdout-limit 16384 for Runtime business commands so native tools do not turn larger inline JSON into a truncated preview before the CLI can return its own receipt.
  • If the native tool still persists or clips stdout (including spill failures), read its complete result file before parsing; never restart the billed request.
  • Small results stay inline: use their full data without a second file read. Read detailed references only for the operation or error you need.
  • With --output, read the full result from meta.output_path; stdout is a brief receipt (data=null). Commands and fields: commands.md.
  • Without --output, results over --stdout-limit bytes (default 65536; LEV8_STDOUT_LIMIT; 0 disables) are saved automatically. Read the complete result from meta.output_path before claiming to have used it; the summary is not the result. Stdout then carries meta.output_path, meta.output_auto=true, meta.output_bytes and meta.output_summary. If meta.output_spill_error is set, the full result is on stdout: use that result without rerunning a billed operation.
  • Without --json, output is a human summary. Save long-term results with --output; upstream Entity retention is finite.

Receipts and recovery

  • Each billing POST saves a private receipt first. After a disconnect, timeout or interruption, take meta.receipt_id and run:
lev8 requests show <receipt-id> --json
lev8 requests resume <receipt-id> --wait --json --output result.json
  • Never generate a new key for a missing response. Resume keeps the original origin, token, body and key; never switch a receipt's token or origin.
  • If meta.result_unknown=true, preserve the record and report the uncertainty. Only retry_with_new_key=true makes a new paid attempt safe, and the user must still authorize it.
  • Ctrl-C and waiting timeouts stop local waiting only. Only synchronous contact can be cancelled: lev8 contact cancel --receipt <id> --wait --json.
  • Receipts hold full request bodies; lev8 requests prune --older-than 30d keeps active and unknown ones. Replay rules: receipts-and-recovery.md.

Exit codes

Exit Meaning Action
0 Success Check completed; not_found is valid
1 Business failure, maybe partial Keep partial data; outcome=error fails at HTTP 200
2 Invalid input, receipt or key conflict Fix input; never change the key silently
3 Token, auth or keyring failure Ask the user; credential_exists needs their choice
4 Credits or concurrency limit Check retry_with_new_key
5 Timeout, transport or unknown state Keep the receipt; resume with the same key
6 Local persistence or output failure Check request_sent and the receipt
7 Device authorization pending Run auth device complete --id later
130 Local interruption Server not cancelled; resume the receipt

A nonzero exit does not prove no billing occurred; report completion, partial results, known costs and unresolved state separately. Details: exit-codes-and-errors.md.

Version History

  • d8ebc34 Current 2026-11-14 00:26

Same Skill Collection

skills/builtin/canvas/1.0.0/SKILL.md
skills/builtin/canvas/1.0.1/SKILL.md
skills/builtin/canvas/1.0.2/SKILL.md
skills/builtin/collaborate/1.0.0/SKILL.md
skills/builtin/create-meeting-minutes/1.0.0/SKILL.md
skills/builtin/docx/1.0.0/SKILL.md
skills/builtin/holistic-context/1.0.0/SKILL.md
skills/builtin/humanlike-browse/1.0.0/SKILL.md
skills/builtin/humanlike-browse/1.0.1/SKILL.md
skills/builtin/humanlike-browse/1.0.2/SKILL.md
skills/builtin/humanlike-browse/1.0.3/SKILL.md
skills/builtin/humanlike-browse/1.0.4/SKILL.md
skills/builtin/humanlike-browse/1.0.8/SKILL.md
skills/builtin/impeccable/1.0.0/SKILL.md
skills/builtin/impeccable/1.0.1/SKILL.md
skills/builtin/lev8-search/1.0.0/SKILL.md
skills/builtin/media-generation/1.0.3/SKILL.md
skills/builtin/pdf/1.0.0/SKILL.md
skills/builtin/pptx/1.0.0/SKILL.md
skills/builtin/pptx/1.0.1/SKILL.md
skills/builtin/pptx/1.0.2/SKILL.md
skills/builtin/proactive-response/1.0.0/SKILL.md
skills/builtin/proactive-response/1.0.1/SKILL.md
skills/builtin/skill-installer/1.0.0/SKILL.md
skills/builtin/slide-deck/1.0.0/SKILL.md
skills/builtin/slide-deck/1.0.1/SKILL.md
skills/builtin/superpowers/1.0.0/SKILL.md
skills/builtin/taste/1.0.0/SKILL.md
skills/builtin/taste/1.0.1/SKILL.md
skills/builtin/teamhub-files/1.0.0/SKILL.md
skills/builtin/teamhub-files/1.0.1/SKILL.md
skills/builtin/teamhub-files/1.0.2/SKILL.md
skills/builtin/teamhub-files/1.0.3/SKILL.md
skills/builtin/xlsx/1.0.0/SKILL.md
skills/builtin/artifact/1.0.0/SKILL.md
skills/builtin/artifact/1.0.1/SKILL.md
skills/builtin/artifact/1.0.2/SKILL.md
skills/builtin/chat-card/1.0.0/SKILL.md
skills/builtin/personal-profile/1.0.0/SKILL.md
skills/builtin/personal-profile/1.0.1/SKILL.md
skills/builtin/personal-profile/1.0.2/SKILL.md
skills/builtin/personal-profile/1.0.3/SKILL.md
skills/builtin/personal-profile/1.0.4/SKILL.md
skills/builtin/personal-profile/1.0.5/SKILL.md

Metadata

Files
0
Version
d8ebc34
Hash
3f9bf368
Indexed
2026-11-14 00:26

inicio - Wiki
Copyright © 2011-2026 iteam. Current version is 2.156.0. UTC+08:00, 2026-11-14 06:03
浙ICP备14020137号-1