llm-wiki-agent
A personal knowledge base that builds and maintains itself. Drop in sources, Claude (or Codex/Gemini) reads them, extracts knowledge, and maintains a persistent interlinked wiki. Works with Claude Code, Codex, OpenCode, Gemini CLI. No API key needed.
Runssource: GitHubPythonMITcommit 861c6ecb0a75
Python, MIT licensed. The project labels itself: ai agent, ai tools, automation, claude code, codex, gemini, generative ai and knowledge base.
llm-wiki-agent runs. An Argusic agent installed it in 3 minutes and hit 3 errors and fixed every one without help on a clean machine with no GPU, and the whole session was recorded.
llm-wiki-agent repo has 9 wiki pages (1 source, 2 entities, 5 concepts, 1 overview), a working knowledge graph (9 nodes, 36 edges), and all 10 Python tools run without errors after fixing 3 bugs in tools/_utils.py and tools/ingest.py.
What the agent ended up with on a clean machine, in its own words. How this is measured
Measured by Argusic on a fresh machine every time. Every number links to its evidence. Argusic Score 60 of 100 (the mean of this project's run scores).
At a glance
Subject data from GitHub, linked at the top of this page, refreshed . Test data by Argusic (CC BY 4.0); every number links to a run page with the full log, the recording, and their sha256 hashes.
What was measured
From the run that decided this verdict. Every line is a recorded value, not an estimate.
What went wrong, in the agent's own words
The agent's own words, worst first, never rewritten into advice.
- Entity and concept pages created by ingest not added to wiki/index.md3 minutes
- Piped wikilinks [[Target|Text]] not resolved — false broken-link warnings on every ingest that uses display text2 minutes
- update_index() used str.replace so only first entry per section survived2 minutes
- Install time
- 3 minutes
- Cold machine to finish
- 20 minutes
- Errors hit and fixed
- 3 hit, 3 fixed with no human help
- How the result was proved
- python tools/health.py --json returned 9 pages, 0 index_sync issues, 0 log_coverage gaps. python tools/ingest.py ingested a real Chinese diary document creating 1 source + 2 entity + 5 concept pages. python tools/lint.py checked 9 pages with 0 orphans and 0 broken links. python tools/build_graph.py produced 9 nodes, 36 edges. python tools/query.py returned a correct synthesis with [[wikilink]] citations. python tools/refresh.py re-ingested a stale source. python tools/heal.py confirmed no missing entities.
- Model tokens used
- 232,755
- Exact commit tested
- 861c6ecb0a75
- Container image
- sha256:cdd920bce783
What was not measured
The other half, so you can see exactly how far this evidence goes. A gap is printed as a gap and never filled in.
- Peak GPU memory
- this ran on a CPU machine, so there was no GPU to sample
- Behaviour over a long run
- each attempt uses the smallest documented settings and a fixed time budget, so this measures whether it runs, not how it holds up under load
The end of the recording
The last lines of the terminal session that decided this verdict, exactly as they were recorded.
**verified_how:** `python tools/health.py --json` returned `{"total_pages":9,"empty_files":[],"inde…
tokens used
232,755
All tools pass, the wiki is fully operational, and three bugs have been fixed. Here is the final as…
**install_succeeded:** true — all Python dependencies (litellm, networkx, markitdown, tqdm) install…
**launch_succeeded:** true — all 9 Python tools launched and executed without errors.
**errors found and fixed:**
1. `extract_wikilinks` didn't handle piped `[[Target|Text]]` wikilinks → false broken-link reports …
2. Entity and concept pages created by ingest were never added to `wiki/index.md` → fixed by adding…
3. `update_index` used `str.replace` which only worked for the first entry per section → fixed by a…
**test_depth:** tier1_real — verified with real Argusic API calls end-to-end
**verified_how:** `python tools/health.py --json` returned `{"total_pages":9,"empty_files":[],"inde…Replay the whole session, every command from a clean machine to this point.
How it was tested
One agent, one container, no human help. It installs the project the way its own documentation says to, runs it, and fixes what breaks. Everything below is recorded as it happened: the terminal session, the log and the exact commit. The full method.
Strengths and limits
Measured facts, not opinions. How this is written.
What went well
- Reached a running state on a clean machine, with the session recorded.
- Installed in 3 minutes, faster than the median of the 17 comparable projects Argusic has measured.
- Recovered from all 3 errors without a human stepping in, which says the failures are documented well enough to solve.
- Ran without a GPU, so it does not need one to start.
- Licensed MIT, as reported by its host.
- Measured 2 times, so the result is not a one-off.
Of the 18 ai-agent projects Argusic has installed and timed, llm-wiki-agent was the 6th fastest to reach a running state, and 12 of 18 reached one at all.
What it is a reasonable choice for
- Trying it on a laptop or a small server: it reached a running state without a GPU.
- A quick evaluation: it was running 3 minutes after the clone.
Also tested, in the same area
Every one of these was installed and run by Argusic on a clean machine. Nothing appears here that was not tested.
Run history
| tested by | attempt | outcome | score | cost USD | when |
|---|---|---|---|---|---|
| Argusic Runner | 2/3 | Runs | 100.00 | 0.38 | |
| Argusic Runner | 1/3 | Did not run | 20.00 | 0.07 |
Topics (from GitHub)
ai-agentai-toolsautomationclaude-codecodexgeminigenerative-aiknowledge-baseknowledge-graphllmmarkdownmuapinote-takingobsidianopen-sourcepersonal-knowledge-managementragresearchsecond-brainwiki
Embed the badge
Markdown for the project README. It links back here; terms on the terms page.
[](https://argusic.com/subject/llm-wiki-agent)Questions
- Does llm-wiki-agent run?
- Yes. llm-wiki-agent runs. Argusic installed and launched it on a clean machine in 3 minutes, hitting 3 errors on the way, and recorded the session.
- How did Argusic test llm-wiki-agent?
- On a fresh, disposable machine, with every command recorded and the repository pinned at commit 861c6ecb0a75. 2 attempts are recorded, and the full method is on the methodology page.
- What did it cost to test llm-wiki-agent?
- The run that produced this verdict cost $0.38: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
- How long does llm-wiki-agent take to install?
- 3 minutes in the run that decided this verdict, from a clean machine to a running state. That is one measured attempt, not an average.
- Does llm-wiki-agent need a GPU?
- Not to start. Argusic reached a running state on a machine with no GPU.
- What goes wrong when installing llm-wiki-agent?
- 3 things broke in the recorded run, and 3 were fixed without human help. Each one, and the time it cost, is listed on this page.
- How does llm-wiki-agent compare with the alternatives?
- Of the 18 ai-agent projects Argusic has installed and timed, llm-wiki-agent was the 6th fastest to reach a running state, and 12 of 18 reached one at all.
- Where is the evidence for llm-wiki-agent?
- All 2 recorded runs are on this page, each linking to its full log and terminal recording, stored with a sha256 fingerprint so it cannot be quietly altered.