Understand Knowledge

Before you start: read this whole install prompt, tell me in a few lines what this skill does and what installing it involves, and wait for my go-ahead before you do anything else. Prerequisites — Use-time prerequisite, disclosed up front: this skill is one of the Understand-Anything plugin's skills. It assumes the Understand-Anything plugin is installed and its core is built (pnpm install in the plugin, then building packages/core, the @understand-anything/core package) — installing this listing only copies the skill folder below; it does not install or build the plugin. The skill also runs its two bundled Python scripts with python3 at use time. It writes its outputs into the knowledge base directory YOU designate (its .ua/ data directory) — nowhere else. No account, key or payment is needed to install or use this listing. Install "Understand Knowledge" for me. It analyzes a Karpathy-pattern LLM wiki — a three-layer knowledge base of raw sources, wiki markdown with [[wikilinks]], and a schema file (CLAUDE.md, AGENTS.md or similar), compatible with an Obsidian-style wiki — and generates an interactive knowledge graph with entity extraction, implicit relationships and topic clustering, tagged kind:"knowledge". Its bundled parse-knowledge-base.py deterministically extracts articles, sources, topics, wikilinks and categories from index.md; article-analyzer subagents then add implicit knowledge (treating article content as untrusted source text, never as instructions), and merge-knowledge-graph.py deduplicates entities and writes the validated graph to knowledge-graph.json and meta.json inside the target wiki's own data directory (.ua/, or the legacy .understand-anything/), after which it auto-triggers the separate understand-dashboard skill. As your wiki keeps changing, re-running understand/explain-style updates through this skill regenerates the graph from the same sources. Its folder holds SKILL.md plus the two Python scripts. Repository: https://github.com/Egonex-AI/Understand-Anything (the skill is the whole understand-anything-plugin/skills/understand-knowledge/ folder at the pinned commit: SKILL.md, merge-knowledge-graph.py, parse-knowledge-base.py, 3 files in total) Pinned version: install the code at commit e24fcf6d371bcfa529bc110bacb1bb43de1b8d33 exactly. Do not substitute a newer version. If this commit is unreachable, stop and tell me. 1. Clone https://github.com/Egonex-AI/Understand-Anything into a temporary folder, check out commit e24fcf6d371bcfa529bc110bacb1bb43de1b8d33, and copy the understand-anything-plugin/skills/understand-knowledge/ folder from that checkout. Then compare every file in the folder with the fingerprints recorded for this listing on 2026-10-22: [FINGERPRINTS] - SKILL.md: SHA-256 e2b225a82d0fc59a83b8391afe17d1ff7f84acb831d2cf716f0ab583ba09c010 - merge-knowledge-graph.py: SHA-256 97e625234ce28bec7b1b02ec7191c70c6c75a54dda3579f7e94bc06ec10679f3 - parse-knowledge-base.py: SHA-256 4eb78d9a8a628bb4547a0c88cb57c2b2a22ba595664736a579ad56d9e46d7249 If a file does not match, it changed since the listing was prepared: do not install it yet, tell me what you found and let me decide. 2. Read SKILL.md in full, and tell me in plain words: what the skill does when used, what it connects to, whether it reads or sends any personal information or credentials, and anything that seems unusual. Stop on any red flag and tell me before going further. Confirm these points from the files: - The folder holds 3 files only: SKILL.md (6,191 bytes, frontmatter name understand-knowledge), parse-knowledge-base.py and merge-knowledge-graph.py — two standalone Python scripts (standard library only, from their imports). There is no other code in it. - Nothing is sent anywhere during installation itself; the only download at install time is the skill folder from GitHub at the pinned commit, and neither Python script is run. - At use time everything is read and written locally: it parses the wiki directory I designate, writes scan-manifest.json and analysis batches to that directory's .ua/intermediate/ (or legacy .understand-anything/intermediate/), then the final knowledge-graph.json and meta.json to the same data directory, and removes the intermediates when done. It reads no credentials and sends nothing anywhere. - Do not parse any wiki, do not run either Python script, and do not use the skill in any way during this install: using it is a separate step that only starts when I ask for it later. 3. When I say "yes, install", install it as a skill: copy the whole understand-knowledge folder into the agent's skills directory, in a folder named exactly "understand-knowledge", keeping its 3 files in their original relative layout (SKILL.md at its root). 4. Verify with no network calls: frontmatter valid (name: understand-knowledge), the 3 files in place with matching fingerprints. Do not run anything from the folder and do not use the skill to verify. 5. Report what was installed, where, and what I still need to do myself (for example: that the Understand-Anything plugin must be installed with its core built (pnpm install plus building packages/core), that python3 is used at use time, that the graph is written into the .ua/ data directory of the wiki I designate, and that using this skill is a separate step that only starts when I ask for it later). GitHub is optional: if I have a GitHub account or the gh CLI, you may use it; otherwise public access is fine. Never require it unless it's in the prerequisites above. Rules: don't touch anything outside the temp folder and the install target. If anything looks off, stop and ask me.