# LLMS-Explorer — vocabulary > Terms of this niche, one per line: canonical name, definition, how it differs from its neighbours (`not:`), and the words people use instead (`aka:`). Read before the index when a term is unfamiliar. 79 terms, 16 with a definition; each line ends in the URL its definition came from. ## Terms - **LLMS-Explorer** — Donate — LLMS-Explorer is free — reading, the tree, every served llms file, no account needed — a one-time or monthly donation keeps it running. — https://llms-explorer.com/donate/ - **Scope** — API keys — Create, list and revoke the scoped keys that authenticate the hosted MCP endpoint; the plaintext is shown once, at creation, and stored only as a hash. · not: Source · differs: vs proprietary llms files — A proposed governance design for conceptual llms files compares scoped claims, retains disag — https://llms-explorer.com/keys/ - **Attribute** — The attribute rubric — Every attribute an llms file is judged on, with bars and severities. — https://llms-explorer.com/reference/attributes/ - **llms.txt** — llms.txt: the spec and its grammars — Spec v2, llms-full grammars, discovery, consumers. · aka: /llms.txt, /llmstxt · not: Change · differs: vs V1 — Compare the llms.txt proposal’s August 2026 v2 changes with the hub pipeline’s V2 artifacts, migration steps, an — https://llms-explorer.com/reference/spec/ - **manifest.json** — Origin allowlist — `manifest.json` declares explicit `host_permissions` for every external surface the extension contacts. There is no `` permission. — https://llms-explorer.com/blog/mdb-case-assistant-project-pitch/ - **Rule** — Reasoning: why the rules are what they are — Extractive descriptions, the size ladder, anchors, facts as the trusted layer, and the two-hop bar — each rule traced to the evidence that produced it. — https://llms-explorer.com/reference/reasoning/ - **Date** — Why we passed on the faster candidates — Passing a tiny tool probe was an entry check. Longer tasks exposed different failures. — https://llms-explorer.com/blog/getting-gemma-mlx-to-do-real-research/ - **sourcesCount** — 8.4 Source-count is a proxy, not a guarantee — `sourcesCount` measures how many citations `/dr` gathered, not whether the synthesis is correct. · not: Measure · differs: not a guarantee — `sourcesCount` measures how many citations `/dr` gathered, not whether the synthesis is correct. — https://llms-explorer.com/blog/a-closed-loop-system-for-autonomous-skill-knowledge-acquisition/ - **/context.md** — Read a concept — Fetch `/context.md` to choose a concept, then fetch its `/downloads/concepts/.md` file. Each file includes its summary, facets, facts and related concepts. · aka: context.md — https://llms-explorer.com/reference/context-files/ - **Content** — The process — Sign in, provide a subject name and content (or upload a file), and submit. A research pass generates an initial summary and fact sheet. — https://llms-explorer.com/contribute/ - **Sample** — Run identity — `run_id = `. Every input is pinned by content hash, so a run is reproducible — and comparable to another run *only if* both share the same sample set and the same resolution version. — https://llms-explorer.com/blog/case-diagnosis-backtest-scoreboard-project-briefing/ - **Study** — Case Study — Gamifying Incident-Response Training with a Rate-Limiting Fire Drill — A training case study of an Atlas API rate-limiting fire drill: how its scorecard rewards restraint, and why team support multiplies, but cannot replace, individual expertise. — https://llms-explorer.com/blog/case-study-gamifying-incident-response-training/ - **Claims** — Saturation, measured — Each pass's claims are diffed against the accumulated list from every prior pass. The new-information rate — new atomic claims this pass ÷ total claims after this pass — is recorded per pass, and the run stops after two consecutive passes each score below 5%. · not: Scope · differs: vs proprietary llms files — A proposed governance design for conceptual llms files compares scoped claims, retains disag — https://llms-explorer.com/skills/rabbithole/ - **llms-vocabulary.txt** — The vocabulary file — How llms-vocabulary.txt defines family terms, labels model-written definitions and supports aliases, with a clear boundary between shipped behavior and proposed senses. — https://llms-explorer.com/blog/vocabulary/ - **Grammar** — llms.txt: the spec and its grammars — Spec v2, llms-full grammars, discovery, consumers. · not: llms-full.txt · differs: not in the spec, and three grammars — `llms-full.txt` (the whole docset inlined into one markdown file) appears nowhere — https://llms-explorer.com/reference/spec/ - **Passes** — The passes — What the optimizer runs, in order, and how each pass is judged and fixed. — https://llms-explorer.com/reference/passes/ ## Named, not yet defined Terms the pool names (≥ 2 uses) without a sourced sentence that defines them — research gaps. - Applies - Bar - --check-links - split_llms_full - verified-as-of · aka: verified-as-of: - Link: rel=describedby · aka: Link: rel="describedby" - text/plain - rel="alternate" type="text/markdown" · aka: rel=alternate type=text/markdown - X-Markdown-Tokens - Effect on an existing file - /.well-known/ -
/llms.txt - page.html.md - page.md - index.html.md - index.md - llms_txt2ctx - V1 (to 2026-08-29) - V2 (from 2026-08-30) - Accept: text/markdown - distill_offline.py bulk - export_llms - web-text-mirror --serve - llms_serve.py - llms_lint.py - docset_rollout cleanup - /ldo - .pages/ - _master.md - ._distill_index.json - PART_PAGES = 60 - --register - section_order - researchedAt - skillId - child concepts · aka: childConcepts - Finding - Window / sample - Claude-Code - Directory - Page - Author / date - Grade - Page block - Source: - /llms.txt - Kind - Measure - Miss - Notes - llms-full.txt - llms-facts.txt - Link - Stage - Change - JSON - Field - Means - Source - Size - Who - llms-small.txt - Bundle