← Back to Skills Marketplace
nextaltair

Diy Pc Ingest

by Altair · GitHub ↗ · v2.0.1
cross-platform ⚠ suspicious
1205
Downloads
0
Stars
0
Active Installs
3
Versions
Install in OpenClaw
/install openclaw-skill-diy-pc-ingest
Description
Ingest pasted PC parts purchase/config text (Discord message receipts, bullet lists) into Notion DIY_PC tables (PCConfig, ストレージ, エンクロージャー, PCInput). Use when the user pastes raw purchase logs/spec notes and wants the AI to classify, enrich via web search, ask follow-up questions for unknowns, and then upsert rows into the correct Notion data sources using the 2025-09-03 data_sources API.
Usage Guidance
This skill appears to do what it claims: parse pasted PC part text and upsert rows into Notion. Before installing or running it: - Provide a dedicated Notion integration token (NOTION_API_KEY) with only the necessary DB/page access; do not reuse high-privilege tokens. - Be aware the skill will write ~/.config/diy-pc-ingest/config.json when bootstrapping IDs and may read ~/.config/notion/api_key or other env-vars (DIY_PC_INGEST_CONFIG, NOTION_API_KEY_FILE). Inspect that config and remove any secrets before sharing. - Review scripts/notion_apply_records.js and bootstrap_config.js (they are readable JS) to confirm behavior and to see what fields will be sent to Notion. - If you want to limit network exposure, run the skill in a controlled environment or temporarily restrict the integration's access while testing. - Because registry metadata omitted required env/config declarations, treat the omission as an informational mismatch (not necessarily malicious) and confirm you can supply the required NOTION_API_KEY and local config before enabling the skill.
Capability Analysis
Type: OpenClaw Skill Name: openclaw-skill-diy-pc-ingest Version: 2.0.1 The skill is designed to ingest PC parts data into Notion databases, which is a legitimate purpose. However, the `SKILL.md` explicitly instructs the AI agent to use `web_search/web_fetch` for data enrichment. While the stated intent is benign, this capability allows the agent to make arbitrary external network requests, which presents a significant vulnerability for prompt injection or data exfiltration if the agent is compromised or given a malicious prompt. The underlying Python and Node.js scripts (`scripts/notion_apply_records.py`, `scripts/notion_apply_records.js`) are otherwise clean, interacting only with the legitimate Notion API and local configuration files, and do not show signs of intentional malice or unauthorized actions beyond their stated purpose.
Capability Assessment
Purpose & Capability
The skill is clearly designed to parse/classify user-pasted PC part text and upsert rows into Notion — the bundled JS/Python scripts call the Notion API and use the 2025-09-03 data_sources/pages endpoints as described. However, the registry metadata claims no required environment variables or config paths, while the SKILL.md and scripts require a Notion token (NOTION_API_KEY or NOTION_TOKEN and fallback file NOTION_API_KEY_FILE) and local config (~/.config/diy-pc-ingest/config.json). That mismatch is an administrative/information omission but not malicious.
Instruction Scope
SKILL.md confines runtime behavior to classifying/extracting fields from pasted text, optionally enriching via web_search/web_fetch, asking clarifying questions, and then calling scripts/notion_apply_records.js to upsert into Notion. The scripts themselves are deterministic and only perform Notion queries/patches/creates. There is no instruction to read unrelated sensitive system files or exfiltrate data to unknown endpoints (all network calls target api.notion.com).
Install Mechanism
There is no install spec (instruction-only skill) and bundled code is plain JS/Python source. No remote downloads or opaque binaries are pulled during install. The only persistence is writing a local config file under the user's home (~/.config/diy-pc-ingest/config.json) via bootstrap behavior.
Credentials
The skill requires a Notion integration token (NOTION_API_KEY/NOTION_TOKEN) to function and may read/write local files (NOTION_API_KEY_FILE, DIY_PC_INGEST_CONFIG, and ~/.config/diy-pc-ingest/config.json). Those are proportional to the stated Notion-upsert purpose, but the registry metadata did not declare these required env vars or config paths — verify you supply a dedicated limited-scope Notion integration token and manage the local config carefully.
Persistence & Privilege
always:false (no forced always-on). The skill will create/overwrite ~/.config/diy-pc-ingest/config.json when bootstrapping Notion IDs if the config is missing, and it will read token files from ~/.config/notion/api_key if used. This is reasonable for convenience but users should be aware of and control that file creation and the token source. The skill does not modify other skills' configs or system-wide settings.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install openclaw-skill-diy-pc-ingest
  3. After installation, invoke the skill by name or use /openclaw-skill-diy-pc-ingest
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v2.0.1
- No functional or documentation changes in this release. - Version bump only; content remains identical to the previous version.
v2.0.0
Version 2.0.0 of openclaw-skill-diy-pc-ingest: - Migrated Notion upsert/apply script to Node.js (`scripts/notion_apply_records.js`) as the new default. - Added support for control fields in JSONL input: `page_id`, `archive`, and `overwrite`. - Introduced optional flag `mirror_to_pcconfig` to allow storage records to also update linked PCConfig rows. - Retained the legacy Python implementation in a `_deprecated` directory for reference. - Updated documentation in SKILL.md for new script, fields, and usage details. - Removed the original Python upsert script from the main scripts directory.
v1.0.1
diy-pc-ingest 1.0.1: Initial release for ingesting DIY PC parts lists into Notion tables. - Parses raw PC parts purchase/configuration text and maps items to the correct Notion DIY_PC tables (PCConfig, ストレージ, エンクロージャー, PCInput). - Extracts and enriches fields using web search when useful, while minimizing unnecessary data filling. - Asks follow-up questions only for ambiguous or required fields to ensure correct data mapping. - Upserts data into Notion using the 2025-09-03 data_sources API, handling creation, update, and safe deduplication based on configurable key rules. - Does not hardcode Notion IDs or tokens; requires user setup via config file.
Metadata
Slug openclaw-skill-diy-pc-ingest
Version 2.0.1
License
All-time Installs 0
Active Installs 0
Total Versions 3
Frequently Asked Questions

What is Diy Pc Ingest?

Ingest pasted PC parts purchase/config text (Discord message receipts, bullet lists) into Notion DIY_PC tables (PCConfig, ストレージ, エンクロージャー, PCInput). Use when the user pastes raw purchase logs/spec notes and wants the AI to classify, enrich via web search, ask follow-up questions for unknowns, and then upsert rows into the correct Notion data sources using the 2025-09-03 data_sources API. It is an AI Agent Skill for Claude Code / OpenClaw, with 1205 downloads so far.

How do I install Diy Pc Ingest?

Run "/install openclaw-skill-diy-pc-ingest" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Diy Pc Ingest free?

Yes, Diy Pc Ingest is completely free (open-source). You can download, install and use it at no cost.

Which platforms does Diy Pc Ingest support?

Diy Pc Ingest is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Diy Pc Ingest?

It is built and maintained by Altair (@nextaltair); the current version is v2.0.1.

💬 Comments