writing-docs
Skill40k starsChanged 39 days agoMITView on GitHub
Writing Quivr's public documentation site. Use when adding or editing a docs page, documenting a new feature, plugin type, API or CLI change, or reviewing documentation.
24 other projects use a version of this.
Trending
More talk lately.thin dataabout the whole repository
Most recently taken up by
- smarzban/agent-skillstheir own version, first seen Nov 15, 2026
- sourcehawk/feature-dev-workflowtheir own version, first seen Nov 15, 2026
- grgrwlkr/claude-dashboard-vscodetheir own version, first seen Nov 8, 2026
- johannschopplich/skillstheir own version, first seen Nov 8, 2026
- Wade-DevCode/awesome-coding-skills-cntheir own version, first seen Nov 8, 2026
- gabrielbauman/skillstheir own version, first seen Nov 1, 2026
Who's talking about The-Vibe-Company/quivr
- Kimz“39.6k Stars:把內容串流變成可監控知識庫的開源引擎”YouTube · Oct 27, 2026 · 4 views
- @golangoss.bsky.social“quivr (⭐️ 39614) An open-source engine that turns continuous content streams into search and monitoring. …”Bluesky · Oct 29, 2026 · 1 likes
- Kimerd“Turn Content Streams Into Searchable Monitored Knowledge”YouTube · Oct 27, 2026 · 1 views
- @golangoss.bsky.social“quivr (⭐️ 39585) An open-source engine that turns continuous content streams into search and monitoring. …”Bluesky · Oct 14, 2026 · 1 likes
- @golangoss.bsky.social“quivr (⭐️ 39603) An open-source engine that turns continuous content streams into search and monitoring. …”Bluesky · Oct 26, 2026
The biggest 5 of 7 mentions linking to the repository that our crawl found. Each line is their own headline.
See all 7
Talked about on Bluesky, YouTube and the web (7 times).
On Bluesky · 3
- Oct 29, 2026quivr (⭐️ 39614) An open-source engine that turns continuous content streams into search and monitoring. Durable ingestion, hybrid search,… — @golangoss.bsky.social1 likes and repost
- Oct 14, 2026quivr (⭐️ 39585) An open-source engine that turns continuous content streams into search and monitoring. Durable ingestion, hybrid search,… — @golangoss.bsky.social1 likes and repost
- Oct 26, 2026quivr (⭐️ 39603) An open-source engine that turns continuous content streams into search and monitoring. Durable ingestion, hybrid search,… — @golangoss.bsky.social0 likes and reposts · 1 reply
On YouTube · 2
- Oct 27, 202639.6k Stars:把內容串流變成可監控知識庫的開源引擎 — Kimz4 views
- Oct 27, 2026Turn Content Streams Into Searchable Monitored Knowledge — Kimerd1 view
On the web · 2
- Oct 30, 2026The-Vibe-Company/Quivr v2.0.1 on GitHub — newreleases.io
- Sep 19, 2023The-Vibe-Company/Quivr v0.0.76 on GitHub — newreleases.io
Used with
Projects with it also use pr-description and release.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
MicrosoftSkillpr-description
Draft Nacelle PR titles, pull request descriptions, commit titles, commit descriptions, and squash commit messages. Use when asked for PR or commit wording, a branch summary, or a PR description that can be used in a commit. Omit validation sections from commit bodies and remove filler.
Also in 3 projects that use this
MicrosoftSkillrelease
Use ONLY when the user explicitly asks to cut/publish a Huabu desktop release or tag a version. Performs an outward action — bumps the version in apps/desktop/package.json, commits + annotated-tags vX.Y.Z, and pushes the branch + tag. That tag push auto-builds the macOS/Windows installers and publishes a GitHub Release. Provides `status`, `bump`, `tag-push`, and a one-command `run` (bump → tag-push).
Also in 4 projects that use this
Other versions24 projects use a version of this.
Word for word · 1
- QuivrHQ/quivr★ 40k100% same
Edited · 0
None found.
Wrote their own version · 23
- remotion-dev/remotion★ 73ksame name
- medusajs/medusa★ 37ksame name
- Nagi-ovo/gemini-voyager★ 21ksame name
- nukeop/nuclear★ 19ksame name
- ngrx/platform★ 8.3ksame name
- cuthbertLab/music21★ 2.6ksame name
- c15t/c15t★ 2ksame name
- NVIDIA/Megatron-Energon★ 388same name
- habit-hooks/habit-hooks★ 273same name
- Wade-DevCode/awesome-coding-skills-cn★ 7same name
13 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository46 other files it gives its agents.
- AGENTS.mdAGENTS.md
- quivr-search/AGENTS.mdAGENTS.md
- sdks/AGENTS.mdAGENTS.md
- armada-coordinatorSkill
- armada-runtime-claude-codeSkill
- armada-runtime-conductorSkill
- armada-runtime-herdrSkill
- armada-workerSkill
- ask-mattSkill
- audit-tests-devSkill
- better-uiSkill
- capture-learning-toolsSkill
- codebase-designSkill
- code-reviewSkill
- coordinate-fleetSkill
- design-frontend-devSkill
- diagnosing-bugsSkill
- documentation-writerSkill
- domain-modelingSkill
- emil-design-engSkill
- grillingSkill
- grill-meSkill
- grill-with-docsSkill
- handoffSkill
- humanizerSkill
- implementSkill
- improve-codebase-architectureSkill
- mermaid-diagramsSkill
- mintlifySkill
- plan-prSkill
- prototypeSkill
- researchSkill
- resolving-merge-conflictsSkill
- review-code-devSkill
- setup-matt-pocock-skillsSkill
- ship-pr-devSkill
- tddSkill
- teachSkill
- to-questionnaireSkill
- to-specSkill
- to-ticketsSkill
- triageSkill
- wait-whatSkill
- wayfinderSkill
- wizardSkill
- writing-for-agentsSkill
The file
We split this skill into the rules it gives an agent
1 rule. 0 are common, 1 is distinctive, and 0 are very specific to this project.
0
Common
Most of the 22 edited versions we read keep these.
1
Distinctive
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 10
Name the reader and their question: Write one line: who arrives on this page (evaluator, operator, plugin author, API client) and what they …
“Common” is measured: kept by at least 6 in 10 of the 22 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
The docs site is the first thing a developer sees of Quivr, and what people use to explain it to others. Every page answers one question for one reader, in plain words and the fewest that fully answer it, with a picture where a picture helps. The bar is the docs of the best-run open-source projects: calm, concrete, and correct to the last command.
Steps
- Name the reader and their question. Write one line: who arrives on this page (evaluator, operator, plugin author, API client) and what they want to do or understand when they leave. Done when the line fits in one sentence.
- Pick exactly one page type: tutorial, how-to, reference or explanation (Diátaxis). Read its template in PAGE-TYPES.md. A page that wants two types is two pages; link them. Done when the page's first heading matches the type's template.
- Find its home in the site map below. Prefer extending an existing page over adding one; search the site folder for the topic first. A new page goes in the navigation next to its siblings. Done when you can name the page before and after it in the navigation.
- Write it following STYLE.md. Use the domain terms of
CONTEXT.md, and explain each in plain words the first time a newcomer meets it. - Run everything you wrote. Every command, request and code block runs on a clean checkout exactly as printed, with its output checked against what the page claims. API requests are runnable blocks (a
{/* runnable */}mark before the fence), whichmake verifyreplays; see STYLE.md for how to write them. A snippet you could not run is removed or introduced as an example ("For example, not run:"). Done when each block has been run once since its last edit, or is introduced as an example. - Regenerate and check the site. After a change to the HTTP contract, a command or the plugin schemas, run
make generate, thenmake docs-site(it rewrites only the generated pages). Then runmake docs,make denylistandmake docs-site-check(Mintlify validation and anchor-checked links), and preview withmake docs-previewany page that uses components.docs/agents/documentation.mdnames every failure rule. Done when all pass and the page renders. - Reread as the reader from step 1, top to bottom, then do the final pass in STYLE.md. Cut every sentence that reader does not need. Done when every line of the final pass holds.
- Run the reader test. Write 5 to 8 questions the reader from step 1 would ask: the page's main answer, a prerequisite or permission, the result of one step, one failure. Give each question to a fresh sub-agent that may read only the page's file, with no other file, tool or conversation. Ask for a short answer, a confidence, and the terms or steps it could not follow, with "the page does not say" instead of a guess. Check each answer against the facts. Fix the page where an answer is wrong or unsure, or where it names a gap that reader needs filled, then ask the failed questions again with new sub-agents. Record the questions, the verdicts and the fixes in the pull request. Done when every answer is right and no needed gap remains.
Site map
The navigation lives in docs-site/docs.json. This table mirrors it; a pull request that changes the navigation updates the table too.
| Tab › group | Holds | Reader |
| Documentation › Get started | Introduction, Quickstart, Core concepts | Someone deciding whether Quivr fits, then trying it |
| Documentation › Plugins | How plugins work, Plugin types, Build your first plugin, and Write a plugin with one how-to per type (push sources included) | Plugin authors |
| Documentation › Run Quivr | Overview, Deploy and configure, Run a large import, Scale Quivr, then one sub-group per task area: Plugins (first-party plugins, pin, upgrade or switch, reprocess quarantine), Sources (receive and secure pushes), Search (search profiles, re-rank with Jev, choose an embedding model, try and switch vector models, fill a vector space), Alerts (alert fields, upgrade a rule, move meaning alerts, retire evaluations), Measure search (measurements, campaigns) | Operators |
| Documentation › Guides | One task per page: add content, search, Alerts (how alerts work, keyword and meaning alerts), follow changes, Collect from sources (connectors, RSS, Microsoft 365, X), connect an AI agent | Integrators calling the API |
| Reference tab | HTTP API overview and endpoints, CLI, MCP tools, plugin manifest and protocol, configuration | Anyone looking up an exact fact |
Name a page's reader from step 1. Plugin-author tasks go in Plugins; operator tasks go in Run Quivr next to the related procedure.
Reference pages generated from contracts (openapi.yaml, reference/cli.mdx, reference/mcp.mdx, reference/plugin-*.mdx) are edited at their source, never in the generated output. A guide joins the site once every step that needs no external credentials has been run end to end. A step that needs a third-party console, a provider account or a paid key may stay: the page lists those credentials in its prerequisites and says which steps were run and which are examples (STYLE.md).
What stays off the public site: contributor process (AGENTS.md, docs/agents/), dated documents and ADRs, internal limits notes. They live in the repository for contributors.
When the product changes
A change to public behaviour updates its page in the same pull request: a new plugin type gets a card on "Plugin types", a how-to and a contract reference entry; a new API field gets its reference and, if it changes a task, the guide that performs that task. Name the page you changed in the pull request.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

