Coordination infrastructure for the people who have to show their work.
Counties, coalitions and recovery networks meet constantly, and most of what gets decided in those rooms is gone within a month. Agora, our first product and the one you can use today, turns a multi-sector planning meeting into a documented record: what was decided, who owns it, and the evidence behind every claim. Four more products are being built on the same privacy architecture, which was written for HIPAA and 42 CFR Part 2 environments from the first line of code.
Coordination breaks in four places, and the meeting is only the first one.
Every county already has the rooms. Public health, the courts, housing, treatment providers, funders and people in recovery all sit down together somewhere. What no county has is the infrastructure underneath: something that turns three hours of discussion into work anyone can act on, that can say which door is actually open tonight, that puts years of documented need in front of a funder, and that keeps track of one person moving through all of it.
The conversation evaporates
Three hours of expert judgment leaves the room as a page of flip-chart bullets and somebody’s memory of who promised what. By the next quarterly meeting nobody can say with any confidence what actually moved.
Nobody knows which door is open
A coalition can map exactly what a person needs and still have no way to find out, at nine at night, which bed or treatment slot will actually take them.
The evidence never reaches the funder
Applications get written from memory in the two weeks before a deadline, while years of need the coalition has already documented sit in a folder nobody thinks to open.
The person falls through anyway
Nobody carries one individual across the same fragmented system the coalition just finished mapping. The organizations that would do the carrying are running on spreadsheets.
These are four missing pieces of infrastructure. Better meeting hygiene does not fix any of them.
Agora runs the whole life of the meeting.
Agora is a browser-based platform for multi-sector community meetings: opioid task forces, recovery coalitions, behavioral health planning bodies. It covers the outcomes agreed before the room convenes and the record left behind afterwards. It is the first of five products and the only one counties are running today, and the core platform has been verified against fixtures taken from real sessions.
Before
Organizers set the event up before anyone arrives: tables assigned by sector, a facilitation framework, invitations, a consent posture. A Desired Outcomes Document states what the meeting is for. The software parses it into structure, but an organizer has to approve that structure before it becomes the reference frame for everything that follows.
During
Each table’s facilitator runs consent-gated, encrypted capture against a transcript they can edit, and it survives network drops. As the conversation moves, an analysis engine surfaces emerging priorities, named barriers and proposed actions for the facilitator to keep or discard. The meeting lead can see every table at once, including the one that has gone quiet.
After
Agora drafts the executive summary, ranks the barriers alongside the sectors that raised them, and removes duplicate proposals. Every claim traces back to the span of transcript it came from. A named person reviews the draft and publishes it. Accepted proposals become commitments with owners, due dates and completion reporting.
This is Agora in one line.
Audio becomes text, text becomes vectors, vectors become named signals, and signals become a draft. No draft becomes a published report without a person releasing it. Every stage in the band below runs in production today.
Illustrative
The privacy constraints are enforced in code.
Agora was written for HIPAA and 42 CFR Part 2 environments from the first line of code, so its constraints are structural instead of procedural. They hold whether or not anyone remembers them. Every product in the portfolio starts from this same floor, which is most of the reason the second one costs less to build than the first.
Nothing captured silently
Capture cannot begin without an explicit consent gate, and a recording indicator stays visible to the room for as long as it runs.
consent gate + recording indicator
Tamper-evident by construction
Every consequential action appends to an audit log where each entry carries the hash of the entry before it. You cannot rewrite history without breaking the chain, and a broken chain is visible to anyone who checks.
sha256 hash-chained audit log
The BAA check is in the code
A vendor without a signed business associate agreement is not a policy violation waiting for someone to notice it. The integration refuses to run.
refuses vendors without a signed BAA
Humans gate every AI output
No model output reaches a participant, a display wall, or a published report without a named person releasing it. There is no auto-publish path.
no auto-publish path
We would rather the system refuse than rely on someone remembering the rule.
Design principle
Phragma gives barriers names, codes and routes through them.
Phragma is a versioned taxonomy: 33 named barrier classes across seven families, each routed to one of nine resolution vectors. Because the vocabulary is fixed and versioned, a barrier raised in one county this spring is comparable to the same barrier raised in another county three years ago. It also looks for shadow barriers, which are the constraints people name out loud in the conversation and then quietly leave out of a table’s formal output.
PH-L1 privacy-regime conflict POR-2 Broker shadow true
The taxonomy itself is built and in use. Automated detection is not. It stays gated behind proven precision on real deployments and a recorded human decision to release it. Until then an analyst does the classifying, and no barrier finding is attributed to a named organization before a person confirms it.
The taxonomy belongs to the portfolio rather than to Agora. When Phoreo tags a barrier at case level, it will use this same versioned vocabulary, so the constraint a navigator hits while carrying one person and the constraint a coalition names in a planning session can be counted as the same thing.
North Carolina’s Exhibit C describes exactly the meeting Agora runs.
To unlock Option B spending, a North Carolina county has to complete a documented collaborative strategic planning process. Holding the meeting is the easy part. Showing afterwards that it was held properly is where counties get stuck.
Flip charts and a summary memo do not meet a documentation standard. Settlement payments run for nearly two decades, so the record a county starts building at its first planning session is the record it will still be reporting against long after everyone in that room has changed jobs.
Payments run into the late 2030s. The documentation obligation runs with them.
The collaborative planning sequence as a working workbook, not a template to fill in afterwards.
No path exists that publishes a report without a named person releasing it.
The planning workbook is built and counties are using it. MOA mode, which maps a county’s own memorandum of agreement onto the same workflow, is still in development.
The longer a community uses Agora, the more its record is worth.
Every published report writes into a longitudinal store scoped to that community: recurring barriers, how priorities have shifted, completion rates by sector. A briefing generator answers the question funders actually ask, which is what this coalition has accomplished and where it is stuck. No claim goes into a briefing without structured evidence behind it, and a published briefing exports as grant evidence, so the meeting held in March becomes part of the application filed in November. The longitudinal store is built. Chaining planning series together across years is still in development.
The record is the asset. Everything we build after Agora reads from it.
Agora is the first of five products, not the whole company.
Convocant Systems is building five products against one coordination problem. Agora documents what a community decides. The other four carry that decision further: to the capacity that has to exist for it, the funding that has to pay for it, the person it was meant to reach, and the organizations doing the work in between. They are a portfolio rather than a product line because each one inherits the spine above instead of rebuilding it.
Poros
Which door is open tonight.
Poros is specified and in a validation phase. Nothing is built yet. It is a recovery-capacity registry and warm-handoff network for the post-clinical layer: recovery residences, faith-based programs, outpatient slots. That is the layer bed registries structurally cannot reach, because those providers are small, largely unlicensed, and answerable to no system that collects their availability.
Convocant Grant Platform
From documented need to funded response.
The grant platform is designed, and one piece of it already exists. It turns a coalition’s published briefings into structured grant-application material, so the application stops being a writing exercise done from memory. The part that is built is the export contract between the two halves: a briefing exports only after a named reviewer has published it, versioned and logged.
Phoreo
Carrying one person across all of it.
Phoreo is on the roadmap and nothing is built. It would handle case-level navigation through the same multi-sector system the coalition coordinates, with barriers tagged against the shared Phragma taxonomy. That shared taxonomy is the argument for building it here rather than somewhere else.
Oikos
The organizations between the meetings.
Oikos is on the roadmap and is the least specified thing on this page. It would be operations infrastructure for the nonprofits that carry the commitments a coalition makes, between the meetings where it makes them. This card is short because padding it out would make it sound further along than it is.
One product is in the field. Of the four here, one is specified and being validated, one is designed with its export contract to Agora already built, and two are commitments with a reason behind them and nothing else yet. Nothing on this page carries a launch date. Every gate in front of these products is a fact about other people’s behavior, and we do not get to schedule those.
Bring us a real meeting. Or tell us we have the order wrong.
We would rather show you Agora against your own agenda, your own sectors and your own reporting requirement than against a demo script. And if what your community needs first is the capacity registry rather than the meeting platform, we would rather hear that from you now than work it out from the roadmap later.