Agentic Software Factory · workshop-2026-07-17

References & Resources

Every tool this workshop demos, mapped to credible alternatives — so you can rebuild each pattern with whatever your machine, budget, or employer allows.

TL;DR

🤖 Have your agent match these to your constraints — paste this into it:

Fetch https://dewy-sleet-n5zs.here.now/agent.md?raw — a machine-readable list of every tool category in this workshop with vetted alternatives (name + URL + evidence). Ask me for my constraints (OS, license policy, budget, what I'm allowed to install), then recommend one alternative per category I can't adopt as-is, with your reasoning.

How to read this page

This workshop demos John's exact stack. You don't need John's exact stack. Every pattern taught here — tiny wrappers, preloaded context, agent-driven browsers, taste voting — works with substitutes. Below: every tool the workshop mentions, what it does in John's system, and credible open-source or commercial alternatives with evidence.

Retrieved 2026-07-17. Star counts, last-push dates, and licenses come from the GitHub API on that date (gh api repos/<owner>/<repo>); non-GitHub products were sanity-checked by web search the same day. "NOASSERTION" means GitHub couldn't map the repo's license to an SPDX id (usually a custom or source-available license) — read the repo's LICENSE before adopting. Nothing here is an endorsement, and a few of John's tools are personal/unreleased — those are flagged, not glossed over (see the appendix).

Terminal coding agents

John uses: pi (71.9k★, MIT) as his primary agent — an AI agent toolkit with a unified LLM API, agent loop, TUI, and an extension system (section 01's pi-session-id is a pi extension; section 06's swarms use pi's subagent tool). He runs Codex CLI (99k★, Apache-2.0), Claude Code (138k★), and Grok side by side behind two-letter zsh wrappers (section 02) — the whole point of that section is that the wrapper pattern is agent-agnostic.

AlternativeWhat it isHow it differsStarsLast pushLicense
OpenCodeOpen-source terminal coding agent, provider-agnostic (any model)Community-run and model-agnostic where pi/Codex/Claude Code are single-author or lab-tied; largest OSS coding agent by stars186,6582026-07-17MIT
Gemini CLIGoogle's open-source terminal agent for Gemini modelsFree tier with a Google account; the Gemini-lab counterpart to Codex/Claude Code106,0242026-07-16Apache-2.0
GooseExtensible AI agent (Block) that installs, executes, edits, and tests with any LLMExtension/MCP-first architecture; works with any LLM rather than one vendor51,2112026-07-17Apache-2.0
AiderAI pair programming in your terminal; git-native diff/commit workflowPair-programming/chat-over-git model rather than an autonomous agent loop; slower release cadence lately (last push 2026-05-22)47,4472026-05-22Apache-2.0
CrushCharm's glamorous terminal coding agent (successor to OpenCode's original TUI)Strongest TUI polish; multi-provider; source-available license (FSL-family, SPDX NOASSERTION)26,5692026-07-17NOASSERTION
OpenHandsAI-driven development platform (formerly OpenDevin) with CLI, GUI, and cloud runtimesHeavier platform (sandboxed runtimes, web UI) vs a single terminal binary81,0452026-07-17NOASSERTION
Grok Build (official) / superagent-ai grok-cli (OSS)xAI's official terminal coding agent (subscription); an OSS coding agent for the Grok APIGrok Build is subscription-gated (SuperGrok/X Premium+) with plan mode and 8 parallel subagents; the superagent-ai CLI is MIT OSS against the Grok API3,2912026-07-06MIT (OSS one); Grok Build is proprietary

Pick this if…

Terminal workspace / agent multiplexers

John uses: Herdr (ogulcancelik/herdr, 17.3k★, AGPL-3.0) — "tmux for agents": panes plus an agent-state sidebar (blocked/working/done/idle auto-detected for 15+ agents), detach/reattach, and — the part section 06 is built on — a unix-socket API that agents themselves can drive: spawn panes, read sibling output, wait on each other. That socket is what makes the keeper prompt ("keep other panes alive") and planned swarms one-keystroke moves.

AlternativeWhat it isHow it differsStarsLast pushLicense
tmuxThe standard terminal multiplexerNo agent-state awareness or agent API; fully scriptable via send-keys/capture-pane so you can hand-roll herdr-like patterns47,8522026-07-16ISC
ZellijTerminal workspace with batteries included (layouts, plugins in WASM)Friendlier defaults and a plugin system, but agent-agnostic — no built-in agent status model34,3302026-07-10MIT
claude-squadManage multiple AI terminal agents (Claude Code, Codex, OpenCode, Amp) in one TUIPurpose-built agent manager like herdr, backed by tmux + git worktrees; narrower pane model8,1312026-06-17AGPL-3.0
Vibe KanbanKanban web UI for orchestrating multiple coding-agent tasksBoard-of-tasks metaphor instead of terminal panes; good for reviewing parallel agent work; last push 2026-04-2427,4052026-04-24Apache-2.0
hcomLets AI agents message, watch, and spawn each other across terminals (Claude Code, Codex, Antigravity, Pi, ...)Agent-to-agent communication layer without owning your terminal layout; closest OSS match to herdr's socket API idea3852026-07-07MIT
OmnaraTalk to your AI agents from anywhere (mobile/web control plane, YC S25)Remote monitoring/answering of agents rather than local pane management; last push 2026-01-192,6502026-01-19Apache-2.0

Pick this if…

Terminal emulators

John uses: Ghostty (58.3k★, MIT) — fast, GPU-accelerated, platform-native. In the stack it's deliberately just the pixel layer; Karabiner scopes its Esc-hold chord layer to the Ghostty app, and Herdr runs inside it.

AlternativeWhat it isHow it differsStarsLast pushLicense
WezTermGPU-accelerated cross-platform terminal emulator AND multiplexer in Rust, Lua-configurableBuilt-in multiplexing could replace part of the herdr layer too; deep Lua scripting27,6362026-07-16NOASSERTION (MIT per repo)
kittyFast, feature-rich GPU-based terminal with 'kittens' extensions and remote control protocolScriptable remote-control protocol (kitty @) is a viable pane-automation substrate33,9132026-07-16GPL-3.0
AlacrittyMinimal, very fast OpenGL terminal emulatorNo tabs/splits by design — pair with tmux/zellij/herdr64,9352026-07-14Apache-2.0
iTerm2The long-standing macOS power-user terminal (Python scripting API, triggers, tmux integration)macOS-only; slower rendering than the GPU generation but deepest macOS integration17,8302026-07-17GPL-2.0

Pick this if…

AI context bundlers (repo → prompt)

John uses: packx (npm install -g packx, johnlindquist on npm) — filters repository files by content string and extension before bundling for AI consumption, with saved search configs. Section 08's oracle skill uses it to bundle up to ~900k tokens per consult. (Related: johnlindquist/pack, a content-filter front-end for Repomix.)

AlternativeWhat it isHow it differsStarsLast pushLicense
RepomixPacks an entire repository into a single AI-friendly file (XML/MD), with token counts and remote-repo supportWhole-repo packing with include/exclude globs; packx's content-string filtering is the main thing it lacks (John's `pack` wraps Repomix to add it)27,1722026-07-15MIT
gitingestReplace 'hub' with 'ingest' in any GitHub URL to get a prompt-friendly extractZero-install for public repos (URL trick + web UI); also a pip CLI15,1332026-07-14MIT
code2promptRust CLI: codebase → single prompt with source tree, Handlebars templating, token countingTemplate-driven output shaping; closest feature set to packx minus content search7,4792026-06-29MIT
files-to-promptSimon Willison's minimal concatenator: directory → one prompt (pairs with his `llm` CLI)Deliberately tiny; no filtering logic beyond paths; last push 2025-02-19 (stable, not active)2,7562025-02-19Apache-2.0
yekFast Rust serializer for repo files, chunks by token budget, priority-orders filesSpeed and token-budget chunking focus; good for very large repos2,4652026-06-29MIT

Pick this if…

Prompt composing (prompt-builder)

John uses: prompt-builder (p, MIT) — a Rust/Ratatui TUI that gives the prompt its own tool before any agent launches: skills lookup ($), file search (@), cross-session history (Ctrl+R, shared with Codex's history.jsonl), then launches pi/Codex/Claude via targets.toml — or hands the prompt as the final argv element to any command (--handoff-command), which is the bridge to mdflow.

Nothing public replicates this exact "compose-then-launch-any-agent" TUI; these cover its axes:

AlternativeWhat it isHow it differsStarsLast pushLicense
fabricOpen-source framework with a crowdsourced library of reusable AI prompts ('patterns') runnable from the CLICurated prompt library + pipe-friendly CLI rather than an interactive composer; patterns ≈ reusable prompt shapes43,1062026-07-16MIT
llm (Simon Willison)CLI for any LLM with reusable templates, fragments, logged history in SQLiteTemplates + logged/searchable prompt history covers prompt-builder's reuse and history axes, minus the TUI and agent handoff12,1942026-07-09Apache-2.0
aichatAll-in-one LLM CLI: Shell Assistant, chat REPL, roles, sessions, RAG, macrosRoles/sessions give reusable prompt setups; REPL-centric rather than compose-then-launch; last push 2026-02-2310,2482026-02-23Apache-2.0
espansoPrivacy-first cross-platform text expander in RustSystem-wide snippet expansion — the low-tech way to get instant prompt scaffolds in ANY agent's input box14,1072026-07-12GPL-3.0
promptfooTest/eval harness for prompts, agents, and RAGs with declarative configsAdjacent, not a composer: it evaluates prompt variants — pairs well with a this-or-that habit23,3432026-07-17MIT

Pick this if…

Markdown agent flows (mdflow)

John uses: mdflow (md, 597★, MIT) — executable markdown prompts run against Claude, Codex, Gemini, or Copilot. The differentiators, per section 05: prompt-build-time command inlining (!`git status`), frontmatter _inputs, isolation from all ambient config/hooks/CLAUDE.md, and eval-gated evolution (md feedback → md eval → md evolve, proposal-first).

AlternativeWhat it isHow it differsStarsLast pushLicense
Claude Code slash commands & skills (built-in)Markdown files in .claude/commands and .claude/skills that become reusable parameterized agent behaviorsShips with the agent — but tied to Claude Code and inherits ambient config; mdflow's isolation and multi-engine targeting are the delta——product docs
dotpromptGoogle's executable GenAI prompt template format (.prompt files: frontmatter + template)A file-format/spec with runtimes, aimed at app code rather than launching coding agents5492026-07-10Apache-2.0
BAMLA programming language for agents/prompts with typed inputs/outputs and testsFull typed DSL + toolchain where mdflow stays plain markdown; heavier but testable8,5602026-07-17Apache-2.0
fabric patternsMarkdown prompt 'patterns' executed via the fabric CLI against any modelSame 'prompt as a file you run' idea, without engine flags, context inlining, or eval-gated evolution43,1062026-07-16MIT
llm templatesYAML templates with params for the llm CLILightweight parameterized prompts; no agent-loop execution or tool use12,1942026-07-09Apache-2.0

Pick this if…

Browser automation for agents (Oracle, agent-browser)

John uses: two distinct things. agent-browser (38.6k★, Apache-2.0) for general "agent drives a browser" tasks. And Oracle (johnlindquist/oracle fork of a private setup) for the section 08 move: copy cookies from your logged-in Chrome profile and drive chatgpt.com / gemini.google.com directly — subscription-priced frontier reasoning (no API keys) and Gemini's video understanding, at ~1,978 sessions of real use. John runs it as oracle serve on a Mac mini behind Tailscale with a dual-account round-robin pool; most people would run it locally.

Honest note: Oracle's cookie-riding consult niche has no maintained public equivalent we could verify. Substitutes are (a) rebuild the pattern on the tools below, or (b) pay for the API: Gemini API / Gemini CLI for video, ChatGPT Pro-tier reasoning via Codex CLI.

AlternativeWhat it isHow it differsStarsLast pushLicense
browser-usePython library making websites accessible to AI agents; can attach to your real Chrome profileGeneral web-task automation with an LLM driver; you could rebuild Oracle's 'drive ChatGPT with cookies' trick on it105,1352026-07-17MIT
Playwright MCPMicrosoft's MCP server exposing Playwright browser control to any MCP-capable agentAccessibility-tree-driven, works out of the box with Claude Code/Codex; persistent profiles carry logins35,1842026-07-15Apache-2.0
StagehandSDK for browser agents (act/extract/observe primitives on Playwright, by Browserbase)Code-first SDK with hosted-browser option rather than a CLI; good for building your own Oracle-like service23,5352026-07-16MIT
SkyvernAutomate browser workflows with LLMs + computer visionWorkflow-oriented (forms, portals) with a self-hostable server; AGPL22,4532026-07-17AGPL-3.0
nanobrowserChrome extension for multi-agent web automation with your own API keysRuns INSIDE your logged-in browser, so it gets cookies for free like Oracle; last push 2025-11-24 (aging)13,4732025-11-24Apache-2.0

Pick this if…

Instant web publishing for agents (here.now)

John uses: here.now — instant hosting for agents: publish any folder to {slug}.here.now (anonymous 24-hour sites with no sign-up; permanent with an API key), plus private cloud Drives agents share across sessions. All 11 workshop pages publish through its skill.

Honest note: the agent-first combination (skill-driven publish + agent Drives + no-signup anonymous sites) is fairly unique. These cover the static-publish half:

AlternativeWhat it isHow it differsStarsLast pushLicense
surge.shCLI for publishing static sites to the surge.sh CDN in one commandThe classic one-command static publish; no agent-native skill or drives3,0442026-07-13none listed
Cloudflare Pages (wrangler)wrangler CLI deploys static sites/functions to Cloudflare's edge (John's inspiration-board uses it)Free, permanent, custom domains; needs an account and a project4,3382026-07-17Apache-2.0
Vercel`vercel deploy` for static or full-stack sites with preview URLsPreview-deploy workflow, generous free tier; heavier platform15,9292026-07-17Apache-2.0
GitHub PagesFree static hosting from a repo branchRequires a git repo (this workshop repo deliberately isn't one); slower publish loop——product
Netlify DropDrag-and-drop a folder, get a live URLClosest no-friction analog for humans; no CLI-first anonymous mode like here.now's——product

Pick this if…

Screenshot capture & annotation (CleanShot X)

John uses: CleanShot X (commercial, macOS) for section 03's whole workflow: hotkey capture → draw a fat red arrow or type the instruction on the screenshot → drag into the agent prompt. The annotation is the prompt.

AlternativeWhat it isHow it differsStarsLast pushLicense
ShottrFast, lightweight macOS screenshot tool: annotate, scrolling capture, OCR, ruler, color pickerPay-what-you-want (from ~$8); no screen recording or cloud; the standard CleanShot substitute for the arrow workflow——freeware/paid (closed source)
FlameshotCross-platform OSS screenshot software with in-place arrow/text annotationFree and cross-platform (Linux/Win/mac); annotation UI is the core feature30,3982026-07-11GPL-3.0
ksnipCross-platform screenshot and annotation tool (Qt)OSS with rich annotation (arrows, numbering, blur); less polished than CleanShot3,2572026-07-03GPL-3.0
macOS built-in Screenshot + Preview markupCmd+Shift+5 capture, then Markup arrows in Preview/Quick LookZero install; slower loop (no capture-annotate-drag single flow)——OS built-in
KapOSS macOS screen recorder (GIF/MP4)Covers the recording half only; effectively unmaintained (last push 2024-11-12) — verify before adopting19,2822024-11-12MIT

Pick this if…

Preference capture / personal RLHF (this-or-that)

John uses: the this-or-that kit (section 10; kit published at gentle-lasso-9d5n.here.now/kit/) — compare.html + a 30-line server.py + options.json: Elo pairing, arrow-key duels, both-bad/skip, per-vote notes, light/dark per-mode tracking; every vote appends to votes.jsonl, which agents read back as explicit taste rules. This repo's own theme was chosen by it (130 votes, 3 rounds).

AlternativeWhat it isHow it differsStarsLast pushLicense
FastChat (Chatbot Arena)The platform behind Chatbot Arena: side-by-side battles + Elo ratings for LLM outputsThe industrial version of the same idea (pairwise vote → Elo); heavyweight, model-comparison focused; last push 2026-05-0139,4902026-05-01Apache-2.0
ArgillaCollaboration tool for building high-quality AI datasets, including preference/ranking recordsTeam-scale dataset curation with a server; overkill for a solo taste record but the real thing for RLHF data5,0382026-07-13Apache-2.0
Label StudioMulti-type data labeling tool with pairwise-comparison templatesGeneral annotation platform; pairwise comparison is one template among many27,8542026-07-17Apache-2.0
potatoPortable academic annotation tool (Python), supports pairwise/best-worst schemesLightweight and self-hostable like the kit, config-driven rather than iframe-anything3972026-07-15GPL-3.0
promptfooPrompt/agent eval harness with side-by-side output comparison viewsAutomated + human-graded evals over prompt variants — the 'evals' complement to taste voting23,3432026-07-17MIT

Pick this if…

Keyboard customization (Karabiner-Elements)

John uses: Karabiner-Elements (22.5k★, Unlicense) with his config written as karabiner.edn compiled by Goku (1.4k★, GPL-3.0, last push 2025-08). Section 06's trick: an app-scoped Esc-hold layer in Ghostty fires internal Ctrl+Alt chords that Herdr's [[keys.command]] bindings catch — Karabiner knows the keyboard, Herdr knows the focused pane.

AlternativeWhat it isHow it differsStarsLast pushLicense
kanataCross-platform advanced keyboard remapper (layers, tap-hold) in RustWorks on Linux/Windows/mac; kmonad-style config; no macOS app-scoping as rich as Karabiner's7,6292026-07-15LGPL-3.0
kmonadAdvanced keyboard manager (Haskell), layers and macros, cross-platformThe original layer-centric remapper; config is Lisp-flavored like Goku's edn5,0312026-06-15MIT
HammerspoonmacOS desktop automation scripted in Lua (hotkeys, window management, app events)Hotkeys plus arbitrary automation logic — can replace the 'chord fires a script' half of John's setup15,7402026-07-08MIT
skhdSimple hotkey daemon for macOSJust hotkey→command bindings in a flat config; minimal but no remapping layers; last push 2025-12-098,0412025-12-09MIT
Keyboard MaestroCommercial macOS macro platform (hotkeys, palettes, app-scoped macros)GUI macro builder, one-time license; closed source——commercial

Pick this if…

Secrets for unsupervised agents (1Password op)

John uses: the 1Password op CLI (commercial). AGENDA directive #10: agents are safe to run unsupervised if secrets live in a vault and get injected per-command (op run / op read) instead of sitting in .env files any agent can read or leak.

AlternativeWhat it isHow it differsStarsLast pushLicense
Bitwarden CLI (bw)CLI for the Bitwarden password manager (open clients, optional self-host via vaultwarden)Open-source clients and self-hosting; scripting ergonomics rougher than op run13,2992026-07-17NOASSERTION (GPL/Bitwarden License mix)
InfisicalOpen-source platform for secrets, certificates, and privileged access; CLI injects env varsTeam/infra-grade secret management with `infisical run` (the closest op-run analog); self-hostable28,1142026-07-17NOASSERTION (MIT + enterprise dirs)
Doppler CLIOfficial CLI for Doppler secrets/config: `doppler run -- cmd` injects secrets as env varsSaaS-hosted; the smoothest run-with-secrets DX after op3842026-07-13Apache-2.0
sopsEncrypt secrets inside files in your repo (age/KMS/PGP), decrypt on useSecrets live encrypted next to code — no vault service at all22,5192026-07-13MPL-2.0
pass + ageThe UNIX password store (GPG) and the modern simple file-encryption toolFully local/offline, scriptable, zero SaaS; you assemble the workflow yourself22,8882026-03-20GPL-2.0 (pass) / BSD-3-Clause (age)

Pick this if…

Agent skills: authoring & distribution (skills CLI)

John uses: the skills CLI (26.4k★; npx skills add <source> --skill <name>, skills.sh) with canonical skills in ~/.agents/skills symlinked into each agent's skill dir. Section 07 even publishes an installable skill on the workshop page itself.

AlternativeWhat it isHow it differsStarsLast pushLicense
anthropics/skillsAnthropic's public repository of Agent SkillsOfficial first-party skill collection to copy from rather than an installer CLI161,8402026-07-17none listed
superpowersAn agentic skills framework and software-development methodologyOpinionated bundled methodology (brainstorm→plan→implement) vs à-la-carte skill installs256,1472026-07-17MIT
Claude Code plugins/marketplaces (built-in)Native plugin + marketplace system bundling skills, commands, hooks, MCP serversFirst-party distribution channel; Claude-Code-only where skills.sh targets many agents——product
MCP serversModel Context Protocol: distribute agent capabilities as tool servers rather than markdown skillsTools-as-processes vs instructions-as-markdown; complementary, heavier to build——open spec (MIT)

Pick this if…

Just-in-time state capture (markers)

John uses: agent-markers (section 09; pattern page at mindful-mango-accy.here.now) — a <300-line, dependency-free recorder: console/fetch/error/click patched into a ring buffer, a ⚑ Mark keystroke pins one sentence of human testimony to nearby events, and the export renders into a redacted agent-ready debug packet. The differentiator is testimony as a first-class artifact and privacy-by-redaction.

AlternativeWhat it isHow it differsStarsLast pushLicense
rrwebRecord and replay the web: DOM-level session recording libraryFull DOM replay vs markers' redacted event trail; the building block most replay products use19,8822026-07-16MIT
OpenReplaySelf-hostable session replay + product analytics for reproducing issuesAlways-on infrastructure with a server; markers is a one-file, moment-of-surprise capture12,2342026-07-16NOASSERTION (ELv2/MIT mix)
HighlightOSS full-stack monitoring: session replay, errors, logs, tracesMonitoring-platform scale; last push 2026-04-169,3312026-04-16NOASSERTION (Apache-2.0 core)
Sentry Session ReplaySession replay wired to error tracking in SentryError-first: replays attach to exceptions rather than to human testimony44,3082026-07-17NOASSERTION (FSL)
JamChrome extension: one-click bug reports with replay, console logs, network requests, device detailsCommercial, human-in-the-loop like markers — the closest 'press a button at the moment of surprise' product; output targets issue trackers, not agent prompts——commercial (free tier)

Pick this if…

Image generation for design variations

John uses: Codex's built-in $imagegen skill (GPT image model, no API key beyond the Codex subscription) and agy — the Antigravity CLI (1.6k★) with Gemini's generate_image — to produce 9 wildly different mockups per prompt, then hands the winner to a coding agent as a reference image (section 07).

AlternativeWhat it isHow it differsStarsLast pushLicense
Gemini CLIGoogle's terminal agent; image generation via Gemini/Imagen models on the free tierSame Google image models as agy without the Antigravity harness106,0242026-07-16Apache-2.0
ComfyUINode-graph GUI/API/backend for diffusion models, fully localLocal + free at inference; scriptable API makes 9x batch generation easy; needs a GPU and setup121,0792026-07-17GPL-3.0
mfluxMLX-native FLUX and other image models on Apple SiliconRuns the 9x pattern locally on the same Mac you present from; CLI-first2,2112026-06-30MIT
InvokeAIProfessional creative engine/WebUI for Stable Diffusion family modelsPolished local UI aimed at designers rather than agent scripting27,6232026-07-15Apache-2.0
AUTOMATIC1111 stable-diffusion-webuiThe historically dominant SD web UIHuge ecosystem but slowing maintenance (last push 2026-03-02); prefer ComfyUI/Invoke for new setups164,1892026-03-02AGPL-3.0

Pick this if…

Also mentioned in the workshop

These came up but don't need an alternatives table:

Appendix

What we could NOT verify (flagged, not fabricated)
  • cs — no public repo found; treated as personal/unreleased, with the honest substitutes listed in §10.
  • packx — verified as John's npm package via web search (npm install -g packx), but it has no public GitHub repo we could find; star/license evidence unavailable.
  • Oracle's exact consult architecture — the fork repo exists (johnlindquist/oracle) but is empty of description/stars; capabilities described here come from the workshop's own docs.
  • Grok Build star data — proprietary, no public repo; existence verified via x.ai announcement coverage.
  • License nuances behind every "NOASSERTION" — read the repo's LICENSE before commercial adoption (notably Crush, OpenHands, herdr, Sentry, OpenReplay, Infisical, Bitwarden).
Provenance & method

gh api repos/<owner>/<repo> (stars, pushed_at, license.spdx_id, description), gh search repos, and WebSearch for non-GitHub products. Star counts and last-push dates recorded verbatim from GitHub API on the retrieval date. 'NOASSERTION' means GitHub could not map the license to an SPDX id (custom/source-available license); check the repo.

Scope: Every tool mentioned in AGENDA.md and the per-section CLAUDE.md files of the workshop repo, with 3-6 verified alternatives per category where they exist. Prose and fit guidance come from the workshop's own materials and the research draft; each category names the section of the workshop it supports. Alternatives were selected for credibility (activity, adoption, licensing clarity) — inclusion is evidence of existence and fit, not endorsement.