solo.ist/coopertron
1
0 Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

synthesize: anonymous braindump CSV → themed review doc via the co-op gateway

Browse Source 
Stdlib-only pipeline for the pre-meeting braindump: CryptPad Form CSV in,
review.md + synthesis.json out. Two model stages (tag atomic points with
verbatim quotes, cluster into themes/novel/dissent) wrapped in deterministic
Python that verifies every quote against the source and computes counts from
item ids, so the model never writes the final document. Defaults to
tinfoil/glm-5.3 with reasoning_effort=low: without it the model spent the
whole 8000-token budget on hidden reasoning and returned nothing (probed
2026-10-19; findings in README). Refuses non-TEE models without a flag.
Content-hash cache so prompt tuning reruns are free. Offline unittest suite
with a fake gateway; smoke-tested live on 5 synthetic responses for ~$0.03.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
This commit is contained in:
solo.istandClaude Fable 5.1 committed 2026-10-19 02:56:50 -04:00
1 parent a8908c84cf
commit 05f4e88867
12 files changed
+2212 -15
Download Patch File Download Diff File Expand all files Collapse all files

No files matched your search

tests/fixtures/sample.csv
Vendored
+13
View File
@@ -0,0 +1,13 @@
"What drew you to the co-op, and what do you use it for today?","Ideas. What should the co-op build, try, or offer next?",Wants. What would make membership more valuable to you personally?,"Needs. What is missing, broken, or in your way right now?",Goals. Where should the co-op be a year from now?,Priorities (pick up to 3),"Dissent. What worries you, what do you disagree with, or what are we getting wrong?",Anything else?
"Privacy. I wanted an AI I could use for client work without feeding a big vendor. Mostly chat, some API from scripts.",A shared prompt library members can contribute to. Also a Matrix bot that answers questions about the docs.,A cheaper tier for light users; I barely touch my allowance most months.,The chat loses my place on mobile when I switch apps. Also the model picker names are confusing (dotted vs dashed slugs).,"Fifty members, real elections, and at least one model we host ourselves.","privacy, pricing and fairness, owning our own infrastructure",I worry we are spreading across too many providers. Three providers is already hard to explain to a new member.,
Green energy mostly. I teach and wanted something I could recommend to students with a straight face.,"Classroom or group accounts: one payment, several seats, shared budget.","Being able to see the energy and emissions per conversation, not just tokens.",Onboarding email landed in spam for two of my students. The email-must-match rule tripped one of them.,"A recognized option for universities and nonprofits, with an institutional tier.","green energy, docs and onboarding, governance and participation","Not dissent exactly, but the Loomio threads are long and I cannot keep up.",Thank you for doing this.
Curiosity about co-op governance. I use the API from a small writing tool I maintain.,An institutional or app-partner membership so projects like mine can buy inference for their users through the co-op instead of each user joining separately.,Stable model names. When tinfoil/glm-5-3-flash disappeared my app broke.,No deprecation notice period for models. No status page.,"A documented SLA-ish commitment, even an informal one, and a changelog members can subscribe to.","API and developer tooling, model choice, privacy",I think the terms forbidding any redistribution of access will block exactly the kind of mission-aligned projects we say we want.,
A friend invited me. I mostly use the chat for drafting emails.,Voice mode on the phone app.,Honestly it is fine as is.,,More members so it keeps existing.,community and chat,,
Wanted to support a cooperative alternative. I use it daily for coding help.,Self-host at least one open model on our own hardware so we are not just reselling other people's APIs.,Higher allowance for people who pay more; the flat $15 cap makes paying extra feel pointless.,Tool calling is flaky on some models and there is no doc saying which ones support it.,"Our own GPU box, even a small one, and a path to more.","owning our own infrastructure, model choice, API and developer tooling",Self-hosting will cost more than buying tokens and we should be honest about that before voting for it.,Happy to help with ops.
Privacy and the TEE story. Chat only.,Private web search inside the enclave instead of our own SearXNG.,Nothing specific.,The chat is slow to start on voice sometimes.,Keep it small and trustworthy. Growth is not a goal for me.,privacy,I disagree with adding providers that are only policy-private. It dilutes the one thing that made me join.,
I run a small nonprofit and we needed shared AI access we could justify ethically.,Group accounts with a shared budget and an admin who can see spend per seat.,An invoice I can hand to our bookkeeper instead of an Open Collective receipt.,The dashboard shows my usage but not my organization's.,An institutional tier with clear terms and pricing.,"pricing and fairness, governance and participation, docs and onboarding",The pay-what-you-can model is lovely but I suspect it is not covering costs. Show us the numbers.,
Governance nerd. I barely use the models; I joined for the experiment.,Publish a monthly financial and usage report and discuss it on Loomio.,"Real decision rights, not advisory votes.",The charter still says decisions are not binding.,A functioning member-elected board and a budget members approve.,"governance and participation, pricing and fairness",We keep adding features before we have governance. Build the decision process first.,Could we do these braindumps every quarter?
Needed an API for a side project and liked the values.,"An Anthropic-compatible /v1/messages endpoint that is actually documented and tested, so Claude-style clients just work.",Per-key spend limits so one script cannot burn my whole month.,Budget exceeded errors arrive mid-stream with no warning beforehand.,"A developer-friendly gateway people choose on merit, not only on values.","API and developer tooling, privacy",,
Climate reasons.,"Default every new member to the greenest model, not the fastest.",A badge or summary of emissions avoided versus a mainstream provider.,,Carbon accounting published per month.,green energy,Tinfoil is not green and we should say so more plainly on the models page.,
Wanted to try open-weight models without setting up my own box. Mostly chat and a bit of API.,"Image generation in the chat picker, not hidden behind an integration button.",Longer memory across chats.,Image models in the picker caused 500 errors and then got removed with no explanation.,"More models, including vision, at the same price.","model choice, community and chat",I do not think self-hosting is worth it at our size.,
Someone mentioned it at a Metagov call.,"A quarterly braindump like this one, and a public roadmap page built from it.",Knowing what other members are working on; a members directory people opt into.,Matrix invites arrived before I knew what Matrix was. A one-page explainer would help.,"A community that knows itself, not just a service.","community and chat, governance and participation, docs and onboarding",,Nice work on the dashboard.
1 What drew you to the co-op, and what do you use it for today? Ideas. What should the co-op build, try, or offer next? Wants. What would make membership more valuable to you personally? Needs. What is missing, broken, or in your way right now? Goals. Where should the co-op be a year from now? Priorities (pick up to 3) Dissent. What worries you, what do you disagree with, or what are we getting wrong? Anything else?
2 Privacy. I wanted an AI I could use for client work without feeding a big vendor. Mostly chat, some API from scripts. A shared prompt library members can contribute to. Also a Matrix bot that answers questions about the docs. A cheaper tier for light users; I barely touch my allowance most months. The chat loses my place on mobile when I switch apps. Also the model picker names are confusing (dotted vs dashed slugs). Fifty members, real elections, and at least one model we host ourselves. privacy, pricing and fairness, owning our own infrastructure I worry we are spreading across too many providers. Three providers is already hard to explain to a new member.
3 Green energy mostly. I teach and wanted something I could recommend to students with a straight face. Classroom or group accounts: one payment, several seats, shared budget. Being able to see the energy and emissions per conversation, not just tokens. Onboarding email landed in spam for two of my students. The email-must-match rule tripped one of them. A recognized option for universities and nonprofits, with an institutional tier. green energy, docs and onboarding, governance and participation Not dissent exactly, but the Loomio threads are long and I cannot keep up. Thank you for doing this.
4 Curiosity about co-op governance. I use the API from a small writing tool I maintain. An institutional or app-partner membership so projects like mine can buy inference for their users through the co-op instead of each user joining separately. Stable model names. When tinfoil/glm-5-3-flash disappeared my app broke. No deprecation notice period for models. No status page. A documented SLA-ish commitment, even an informal one, and a changelog members can subscribe to. API and developer tooling, model choice, privacy I think the terms forbidding any redistribution of access will block exactly the kind of mission-aligned projects we say we want.
5 A friend invited me. I mostly use the chat for drafting emails. Voice mode on the phone app. Honestly it is fine as is. More members so it keeps existing. community and chat
6 Wanted to support a cooperative alternative. I use it daily for coding help. Self-host at least one open model on our own hardware so we are not just reselling other people's APIs. Higher allowance for people who pay more; the flat $15 cap makes paying extra feel pointless. Tool calling is flaky on some models and there is no doc saying which ones support it. Our own GPU box, even a small one, and a path to more. owning our own infrastructure, model choice, API and developer tooling Self-hosting will cost more than buying tokens and we should be honest about that before voting for it. Happy to help with ops.
7 Privacy and the TEE story. Chat only. Private web search inside the enclave instead of our own SearXNG. Nothing specific. The chat is slow to start on voice sometimes. Keep it small and trustworthy. Growth is not a goal for me. privacy I disagree with adding providers that are only policy-private. It dilutes the one thing that made me join.
8 I run a small nonprofit and we needed shared AI access we could justify ethically. Group accounts with a shared budget and an admin who can see spend per seat. An invoice I can hand to our bookkeeper instead of an Open Collective receipt. The dashboard shows my usage but not my organization's. An institutional tier with clear terms and pricing. pricing and fairness, governance and participation, docs and onboarding The pay-what-you-can model is lovely but I suspect it is not covering costs. Show us the numbers.
9 Governance nerd. I barely use the models; I joined for the experiment. Publish a monthly financial and usage report and discuss it on Loomio. Real decision rights, not advisory votes. The charter still says decisions are not binding. A functioning member-elected board and a budget members approve. governance and participation, pricing and fairness We keep adding features before we have governance. Build the decision process first. Could we do these braindumps every quarter?
10 Needed an API for a side project and liked the values. An Anthropic-compatible /v1/messages endpoint that is actually documented and tested, so Claude-style clients just work. Per-key spend limits so one script cannot burn my whole month. Budget exceeded errors arrive mid-stream with no warning beforehand. A developer-friendly gateway people choose on merit, not only on values. API and developer tooling, privacy
11 Climate reasons. Default every new member to the greenest model, not the fastest. A badge or summary of emissions avoided versus a mainstream provider. Carbon accounting published per month. green energy Tinfoil is not green and we should say so more plainly on the models page.
12 Wanted to try open-weight models without setting up my own box. Mostly chat and a bit of API. Image generation in the chat picker, not hidden behind an integration button. Longer memory across chats. Image models in the picker caused 500 errors and then got removed with no explanation. More models, including vision, at the same price. model choice, community and chat I do not think self-hosting is worth it at our size.
13 Someone mentioned it at a Metagov call. A quarterly braindump like this one, and a public roadmap page built from it. Knowing what other members are working on; a members directory people opt into. Matrix invites arrived before I knew what Matrix was. A one-page explainer would help. A community that knows itself, not just a service. community and chat, governance and participation, docs and onboarding Nice work on the dashboard.