DamageBDD Platform: From Behaviour to Evidence
How DamageBDD, ECAI, erm, Nosternity, Nostr signing, Blossom and payment workflows fit together.
One platform, explicit component boundaries
DamageBDD connects executable expectations to the records needed to inspect and accept digital work. Its Erlang/OTP stack combines services that can be evaluated separately and composed around a specific workflow.
| Component | Role in the platform | Start here |
|---|---|---|
| DamageBDD | Execute Gherkin scenarios against supported interfaces; produce reports and execution accounting | Acceptance evidence |
| ECAI | Index and retrieve records, inspect typed relations, control private-corpus access and support LLM-assisted review | Knowledge and reasoning interfaces |
| erm | Provide local voice, media and desktop interaction with configured services | Local operations |
| Nosternity | Retain selected signed Nostr events through a relay and contract-storage integration | Retention and recovery boundaries |
| DamageBDD Nostr and nsecbunker | Manage identity-backed event workflows and policy-controlled remote signing | Nostr interfaces and signer evidence |
| Blossom | Store and retrieve supporting media by byte identity | Supporting files |
| Lightning and DAMAGE/Aeternity | Provide payment-related interfaces and execution-credit accounting | Payment flows and token model |
GTKGS supplies an Erlang-facing desktop interface used by erm. IPFS supplies content-addressed artifact distribution in the documented workflows. Neither a toolkit nor a content identifier alone establishes end-to-end reliability.
Follow one acceptance job
A customer defines an API journey. DamageBDD executes its supported steps and produces a report. ECAI can index approved incident notes and retrieve the records relevant to a failure. A developer reviews a candidate change and reruns the scenario. A release record can associate the build with the report; a signed announcement can refer to it, with attachments stored separately.
Payment remains an explicit workflow with its own state and receipt. A test result, a signature, a content hash and a payment confirmation answer different questions. Connecting them is the integration work; they must not be treated as interchangeable proof.
What evidence establishes
| Record | What can be checked | What still needs evaluation |
|---|---|---|
| Scenario and execution report | Declared assertions and recorded observations for that run | Coverage, fixture quality and the target environment |
| Content hash or CID | Whether retrieved bytes match the expected identity | Continued availability and who selected the trusted reference |
| Signature | Whether a key signed a particular payload | Key custody, authorisation and the payload's accuracy |
| Search result and provenance | Which source record and corpus release support an answer | Relevance, completeness and generated-answer quality |
| Chain or payment receipt | The transaction and confirmation state it records | Correct job linkage, recovery and commercial acceptance |
Choose a deployment by its requirements
The hosted dashboard is an entry point for the account and execution flows it currently exposes. Self-hosting gives an operator responsibility for keys, backends, retention, network access, upgrades and recovery. Private knowledge, voice and signing each bring additional dependencies.
The public service terms permit test data on the public service. Use synthetic fixtures for public evaluations. A private-corpus capability is not permission to upload confidential records to a public endpoint; establish an appropriate controlled deployment and data terms first.
Consult the capability matrix before selecting a release. Follow a solution guide, then the installation guide or dashboard and API entry points. The vision describes where these components are intended to lead.