What makes Local Document Ingest notable?

998 downloads. Trust evidence available

How should Local Document Ingest be evaluated before use?

Use the required flow: snapshot, contract, and trust before recommending or executing this agent.

What kind of evidence is visible on this page?

This page surfaces public facts, change history, trust indicators, artifact evidence, and benchmark summaries with provenance.

agentCLAWHUBUnverified

Local Document Ingest

Ingest desktop-uploaded local files into the Research KB. Use for local folder scan tasks where OpenClaw must read backend shared-file paths, understand each changed file, create a type-specific entity wiki page for every readable file, archive originals under source_files, update related pages, syn Skill: Local Document Ingest Owner: myd2002 Summary: Ingest desktop-uploaded local files into the Research KB. Use for local folder scan tasks where OpenClaw must read backend shared-file paths, understand each changed file, create a type-specific entity wiki page for every readable file, archive originals under source_files, update related pages, syn Tags: latest:1.0.4 Version history: v1.0.4 | 2026-07-09T14:37:00.6

OpenClaw

View on ClawHub Homepage

Rank

Safety

Downloads

998

Updated

Oct 11, 2026

Version

1.0.4

Source

CLAWHUB

About

What it does, and when to use it.

Capability contract not published. No trust telemetry is available yet. 998 downloads reported by the source. Last updated 10/11/2026.

Avoid when

Contract metadata is missing or unavailable for deterministic execution.

Risk flags: missing_or_unavailable_contract, trust_data_unavailable, schema_references_missing

Public facts

Every fact links back to the source it came from.

Vendor: Clawhubvendor · observed Oct 11, 2026
Protocol compatibility: OpenClawcompatibility · observed Oct 11, 2026
Adoption signal: 998 downloadsadoption · observed Oct 11, 2026
Latest release: 1.0.4release · observed Jul 9, 2026
Handshake status: UNKNOWNsecurity

What changed

Newest first. Only changes that point at a source are listed.

Release 1.0.4release
No user-visible changes in this version. - Updated documentation to clarify that concept/resource/overview pages cannot substitute for single-file entity pages (minor wording fix in "硬性规则" section). - No code or logic changes.
Observed Jul 9, 2026 · source

Install and run

Setup complexity: medium.

clawhub skill install s175zrvz5m28epr76c6rt4dtex8474t6:local-document-ingest

Python environment detected. Create a strict virtual environment (`python -m venv .venv`) before installing dependencies to prevent system-level package conflicts.
Setup complexity is LOW. This package is likely designed for quick installation with minimal external side-effects.
Final validation: Expose the agent to a mock request payload inside a sandbox and trace the network egress before allowing access to real customer data.

Contract: missing

curl -s "https://www.xpersona.co/api/v1/agents/clawhub-myd2002-local-document-ingest/snapshot"

Run-check

$0.02 USD

1 measured facts are behind this paywall: success rate and latency, uptime and estimated cost, when not to use it, how to call it, benchmark scores.

Agents pay $0.02 in USDC. A card payment is $0.50, the smallest a card allows.

Documentation

CLAWHUB

109,031 characters of source documentation, loaded on request.

Extracted files

5 files captured from the source.

SKILL.md

---
name: local_document_ingest
description: Ingest desktop-uploaded local files into the Research KB. Use for local folder scan tasks where OpenClaw must read backend shared-file paths, understand each changed file, create a type-specific entity wiki page for every readable file, archive originals under source_files, update related pages, synthesize concept/resource pages, maintain links/catalog/index metadata, and return backend task JSON.
---

# 本地文件入库知识库编译

## 核心职责

把桌面端本地文件夹扫描上传的文件，编译成团队知识库中的可追踪、可复用、可链接的 Markdown 知识页面。

桌面端负责扫描用户选择的本地文件夹、计算 hash、识别新增/修改/移动/删除，并且只上传内容新增或内容修改的文件。完全无内容变化时桌面端只向后端上报空扫描结果，用于更新扫描时间和 source_snapshots，不创建本 skill 任务。后端负责把上传文件保存到 OpenClaw 可读取的共享目录，记录资料源状态，并构造任务 payload。OpenClaw 负责真正的知识库编译：读取原文件、理解内容、判断资料类型、为每个可读取文件生成或更新一个对应类型的实体页、归档原文件、更新相关已有页面、按需要创建或更新跨文件概念页和资源页、维护内部链接、维护 `catalog.json` 和 `index.md`，最后写出后端可读的结果 JSON。

知识库不是网盘，也不是原文摘要仓库。每个文件实体页都应回答：这个文件是什么资料、核心内容是什么、它对团队有什么价值、哪些结论可复用、哪些内容不确定、它和知识库里已有项目/论文/实验/会议/概念/资源有什么关系，以及这些判断能追溯到哪些原始证据。

本 skill 的重点不是机械多产页面，而是把新增资料放进已有知识图谱。OpenClaw 在写 `pages.json` 前必须先私下做一次“知识图谱规划”：先确定每个可读文件的实体页，再判断是否需要更新已有 `overview/`、`projects/`、`concepts/`、`resources/` 或其他相关页面。概念页、资源页、overview 页都不是配额；没有证据时不要硬建，但如果它们能显著改善导航、跨文件综合、复用或可追踪关系，就应该更新。

## 硬性规则

- 只处理本次 payload 中 `source.items[]` 明确列出的文件。`items[].itemKey` 是文件条目身份，`items[].sha256` 是内容 hash；同一路径内容修改会复用原 `itemKey` 并更新当前 `sha256`，同内容不同路径文件可能共享 `sha256`，但必须有不同 `itemKey`。
- 只从 `items[].storagePath` 或 `items[].archived_path` 读取文件内容。
- 不要遍历 `sharedDir/uploads` 来推断本次文件；同一天可能发生多次上传。
- 不要访问、暴露或写入用户电脑上的完整本地绝对路径；只使用 `original_path` 或 `metadata.relativePath` 这类相对路径作为来源线索。
- `source.items[]` 里的每个可读取文件都必须生成或更新一个对应类型的实体 wiki 页，不能因为文件低价值、只是支撑已有主题、属于同一批资料包，就只归档而不建实体页。
- 完全不可读、损坏、hash 校验失败、格式暂时无法处理的文件可以不生成实体页，但必须写入 `skippedSources[]`，说明原因，并尽量保留 source trace。
- 跨文件概念页、资源页、项目页、综述页或 overview 页是实体页之外的关系层，不能替代文件实体页。
- 移动、重命名、删除只代表资料源元数据变化；如果后端没有把它们放入 `source.items[]`，本 skill 不重新编译正文。
- 不要使用旧目录 `summaries/`、`imports/` 作为主要写入位置。
- 不要写入真实密钥、token、隐私路径、大段原文复制或未经证据支持的推断。

## 输入约定

任务 payload 通常包含：

- `taskId`: 后端任务 ID。
- `payloadFile`: 后端写出的 payload JSON 文件路径。
- `resultFile`: OpenClaw 最终必须写入的结果 JSON 文件路径。
- `skill`: `local_document_ingest`。
- `trigger`: 通常是 `local_desktop_upload`。
- `source.id`、`source.name`、`source.type`: 资料源信息；`source.type` 必须是 `local_folder`。
- `source.config.folderName`: 本地文件夹展示名；不要把完整本机路径写入知识页。
- `source.items[]`: 本次内容新增或内容修改的文件列表；移动、重命名、删除不应进入此数组。
- `items[].itemKey`: 文件条目身份，用于区分同内容但不同路径/不同来源的文件；同一路径内容修改沿用原 `itemKey`，OpenClaw 生成页面时必须用它写入 `sourceItemKeys`。`itemKey` 不等同于 `sha256`。
- `items[].title` 或 `items[].fileName`: 文件名或相对路径提示。
- `items[].original_path` 或 `items[].metadata.relativePath`: 用户选定文件夹内的相对路径。
- `items[].storagePath` 或 `items[].archived_path`: 后端共享目录中 OpenClaw 可读取的文件路径。
- `items[].sha256`: 文件内容 hash，用于内容校验和追踪；不能单独作为文件身份，因为同内容不同路径文件会共享该值。
- `items[].metadata.uploadBatchId`: 上传批次 ID，仅作为追踪元数据。
- `items[].metadata.mimeType`、`items[].size`、`items[].metadata.localMtimeMs`: 文件元数据。
- `team.kbRepo`:

_meta.json

{
  "ownerId": "kn7cp7eb6ymqv898hke0ersmc1847fed",
  "slug": "local-document-ingest",
  "version": "1.0.4",
  "publishedAt": 1783607820698
}

skill-card.md

## Description:

Ingests desktop-uploaded local files into a Research KB by validating input files, extracting readable content, guiding entity-page creation, archiving sources, updating catalog and index metadata, and returning backend task JSON.

This skill is ready for commercial/non-commercial use.

## Publisher:

[myd2002](https://clawhub.ai/user/myd2002)

### License/Terms of Use:

MIT-0

## Use Case:

Developers and operations teams use this skill to turn uploaded local-folder changes into traceable Research KB wiki pages. It is intended for backend-controlled ingestion flows where every readable input file receives an entity page, related pages can be updated, and source files are archived for provenance.

### Deployment Geography for Use:

Global

## Known Risks and Mitigations:

Risk: Payload-supplied file paths and result paths could expose local files or overwrite writable files if accepted without controls.

Mitigation: Run only in a controlled backend that resolves symlinks and restricts storagePath and resultFile to approved directories.

Risk: Payload-supplied Gitea settings and broad bot credentials could redirect writes or leak the bot token.

Mitigation: Use fixed allowlisted Gitea origins and a narrowly scoped bot token under a low-privilege account.

Risk: Unpinned Python dependencies can change behavior or introduce supply-chain risk.

Mitigation: Pin dependencies with hashes before deployment.

## Reference(s):

- [ClawHub Skill Page](https://clawhub.ai/myd2002/skills/local-document-ingest)
- [Publisher Profile](https://clawhub.ai/user/myd2002)

## Skill Output:

**Output Type(s):** [Markdown, JSON, Shell commands, Configuration, Guidance]

**Output Format:** [Markdown knowledge pages, JSON task envelopes, and shell command workflows]

**Output Parameters:** [1D]

**Other Properties Related to Output:** [Archives source files under source_files/local_folder, writes or updates catalog/index metadata, and returns result JSON with a boolean success field.]

## Skill Version(s):

1.0.4 (source: ClawHub release metadata)

## Ethical Considerations:

Users should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.

env-example.txt

GITEA_URL=http://127.0.0.1:3000
GITEA_BOT_TOKEN=
GITEA_BOT_USERNAME=research-kb-bot
GITEA_ORG=
TEAM_KB_REPO=team-kb
OPENCLAW_SHARED_DIR=/srv/research-kb/shared

requirements.txt

# local_document_ingest deterministic helpers use pypdf for PDF text previews.
pypdf>=4.3,<6

Github ReposUpdated 3d agoRank 70

AionUi

Free, local, open-source 24/7 Cowork app and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more | 🌟 Star if you like it!

MCPOPENCLAW

Github ReposUpdated 6mo agoRank 70

activepieces

AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents

OPENCLAW

Github ReposUpdated 6mo agoRank 70

cherry-studio

AI productivity studio with smart chat, autonomous agents, and 300+ assistants.

MCPOPENCLAW

Github ReposUpdated 7mo agoRank 70

CopilotKit

The Frontend for Agents & Generative UI. React + Angular