← Back to Skills Marketplace
lujfsd

gety-local-search

by lujfsd · GitHub ↗ · v1.0.1 · MIT-0
cross-platform ⚠ suspicious
101
Downloads
0
Stars
0
Active Installs
2
Versions
Install in OpenClaw
/install gety-local-search
Description
This skill should be used when the user wants to search for local files or documents on their computer. Trigger phrases include "帮我找文件", "搜索本地文件", "查找文档", "本...
README (SKILL.md)

\r \r

Gety Local Search\r

\r

Overview\r

\r To search local files and documents indexed by Gety, use the gety CLI. Gety is an AI-powered local file search engine that supports semantic search across all indexed content on the user's machine.\r \r

Quick Start\r

\r To perform a basic search:\r \r

gety search "\x3Cquery>"\r
```\r
\r
To get structured JSON output (useful for further processing):\r
\r
```bash\r
gety search "\x3Cquery>" --json\r
```\r
\r
## Core Operations\r
\r
### Searching Documents\r
\r
To search indexed content, use `gety search` with the user's query. Key options:\r
\r
| Option | Description |\r
| --- | --- |\r
| `-n, --limit \x3Cn>` | Max results to return (default: 10) |\r
| `--offset \x3Cn>` | Pagination offset (default: 0) |\r
| `-c, --connector \x3Cname>` | Filter by connector name (repeatable or comma-separated) |\r
| `-e, --ext \x3Cext>` | Filter by file extension (e.g. `pdf,docx`) |\r
| `--match-scope \x3Cscope>` | Filter by match type: `title`, `content`, `semantic` |\r
| `--sort-by \x3Cfield>` | Sort by `default` or `update_time` |\r
| `--sort-order \x3Cdir>` | `ascending` or `descending` |\r
| `--update-time-from \x3Ciso8601>` | Filter by update time (from) |\r
| `--update-time-to \x3Ciso8601>` | Filter by update time (to) |\r
| `--no-semantic-search` | Disable semantic search (enabled by default) |\r
\r
**Examples:**\r
\r
```bash\r
# Basic search\r
gety search "meeting notes"\r
\r
# Search with more results and pagination\r
gety search "roadmap" -n 20 --offset 20\r
\r
# Filter by connector and file type\r
gety search "security review" -c "Folder: Work" -e pdf,docx\r
\r
# Sort by most recently updated\r
gety search "design system" --match-scope title,content --sort-by update_time --sort-order descending\r
\r
# JSON output for processing\r
gety search "project plan" --json\r
```\r
\r
### Fetching a Specific Document\r
\r
To retrieve detailed content of a specific document by its connector ID and document ID:\r
\r
```bash\r
gety doc \x3Cconnector_id> \x3Cdoc_id>\r
gety doc \x3Cconnector_id> \x3Cdoc_id> --json\r
```\r
\r
### Managing Connectors\r
\r
Connectors define which local directories Gety indexes.\r
\r
```bash\r
# List all available connectors\r
gety connector list\r
\r
# Add a new directory to index\r
gety connector add /path/to/folder --name "Folder: Work"\r
\r
# Remove a connector by ID\r
gety connector remove folder_1\r
```\r
\r
## Workflow\r
\r
1. **Receive the user's search request** — identify the query intent (file name, content topic, file type, date range, etc.)\r
2. **Run `gety connector list`** if unsure which connectors exist, to understand the indexed scope\r
3. **Run `gety search "\x3Cquery>"`** with appropriate options based on the user's intent\r
4. **Present results clearly** — show file name, location, and a brief description of matched content\r
5. **Fetch full content if needed** — use `gety doc \x3Cconnector_id> \x3Cdoc_id>` when the user needs to read the document\r
6. **Handle no-results gracefully** — suggest broadening the query, checking connector coverage, or verifying the index is up to date\r
7. **Handle exit codes** — check the process exit code after each command:\r
   - `69`: Gety is not running — ask the user to start the Gety desktop app and retry\r
   - `77`: Quota exceeded — inform the user and suggest retrying later\r
   - `2`: Invalid arguments — check command syntax with `gety search --help`\r
   - `1`: General error — report the error message to the user\r
   - Full exit code reference: `references/gety_cli.md`\r
\r
## References\r
\r
Full CLI documentation is available in `references/gety_cli.md`.\r
Usage Guidance
This skill is a thin wrapper for the third-party 'gety' CLI. Before installing/using it: 1) Verify where the gety binary comes from (official site, package repo) and that you trust that source — the skill metadata does not provide a homepage or install instructions. 2) If you must install gety, prefer official release channels and inspect the binary/package. 3) Be cautious when adding connectors: avoid indexing sensitive directories (password stores, SSH keys, system configs) unless you understand how gety stores and secures indexed data. 4) If you want the agent to run searches, consider running initial gety commands yourself to confirm expected output and exit codes. 5) If provenance cannot be established, do not install or run the CLI — the skill otherwise behaves as described but relies on external software of unknown trustworthiness.
Capability Analysis
Type: OpenClaw Skill Name: gety-local-search Version: 1.0.1 The skill provides a legitimate interface for the 'Gety' local search CLI, allowing an AI agent to search, retrieve, and manage indexed local documents. The instructions in SKILL.md and references/gety_cli.md are consistent with the stated purpose of local file discovery and do not contain evidence of data exfiltration, malicious execution, or harmful prompt injection.
Capability Assessment
Purpose & Capability
SKILL.md describes a local-search skill that operates by invoking a 'gety' CLI — this is coherent with the stated purpose. However the skill metadata does not declare the gety binary as a required dependency, provides no homepage/source, and includes no install instructions; those omissions reduce provenance and make it unclear how the CLI will be obtained or whether it is trustworthy.
Instruction Scope
Runtime instructions stay within the stated scope: list connectors, run search, fetch document content, handle exit codes. They implicitly instruct the agent to run shell CLI commands that will access local indexed files and to add connectors (directories). The instructions do not direct data to external endpoints or request unrelated environment variables, but adding connectors can expose arbitrary filesystem paths to the search engine — the agent will therefore potentially read local content as intended.
Install Mechanism
This is an instruction-only skill with no install spec. That is low risk for disk writes by the skill itself, but the skill assumes an external 'gety' CLI is present and running. Because there is no source/homepage or guidance for obtaining a trusted binary, users may need to install software from an unknown origin — this is a provenance concern rather than direct code-injection from the skill.
Credentials
The skill requests no environment variables, credentials, or config paths. That is proportionate to its purpose. Note that the CLI operations will access local files (as expected) but no extra secrets or unrelated credentials are being requested by the skill itself.
Persistence & Privilege
The skill does not request always:true or other elevated persistence. It's user-invocable and allows normal autonomous invocation. It does not modify other skills or system configuration in the provided instructions.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install gety-local-search
  3. After installation, invoke the skill by name or use /gety-local-search
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.1
- Added workflow steps for handling `gety` process exit codes, including guidance for exit codes 69, 77, 2, and 1. - Updated instructions to inform users what to do if Gety is not running, quota is exceeded, arguments are invalid, or a general error occurs. - Clarified that full exit code references are available in the documentation.
v1.0.0
- Initial release of gety-local-search skill for searching local files and documents using the [Gety](https://gety.ai/) CLI. - Supports semantic and filtered searches across indexed local content. - Enables querying by file name, content, connector, extension, and date range. - Provides options for result limits, sorting, and structured JSON output. - Allows connector management to define indexed directories. - Includes commands for fetching specific document content.
Metadata
Slug gety-local-search
Version 1.0.1
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 2
Frequently Asked Questions

What is gety-local-search?

This skill should be used when the user wants to search for local files or documents on their computer. Trigger phrases include "帮我找文件", "搜索本地文件", "查找文档", "本... It is an AI Agent Skill for Claude Code / OpenClaw, with 101 downloads so far.

How do I install gety-local-search?

Run "/install gety-local-search" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is gety-local-search free?

Yes, gety-local-search is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does gety-local-search support?

gety-local-search is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created gety-local-search?

It is built and maintained by lujfsd (@lujfsd); the current version is v1.0.1.

💬 Comments