Practical overview
When it fits
Use Brainblast before AI-assisted implementation when a project depends on external APIs, SDKs or infrastructure and integration mistakes could remain silent. Run the CLI afterward when those findings need automated enforcement in CI.
Who it is for
- Teams using AI coding agents with third-party APIs or SDKs.
- Developers integrating services such as Stripe, Solana-related systems, authentication providers or webhooks.
- Engineering teams that want research findings represented in reports and CI decisions.
Documented uses
- Research external components before implementation.
- Identify configuration, authentication, payment, webhook and immutable-choice risks.
- Carry research findings into later coding sessions.
- Scan code offline and gate CI on unresolved risks.
- Expose findings through prose and machine-readable reports.
Requirements & limits
The repository is public, lists an MIT license, and documents an npm package used through npx. The evidence does not provide pricing or account requirements.
- The workflow is split between a pre-coding research skill and a CI-oriented npm CLI.
- The default static auditor operates offline without network access or an LLM.
- Research is cached per component and version, with a documented fresh-research option.
- Outputs include final-report.md and a versioned report.json contract.
- CI enforcement can fail on a selected risk severity or a blocked verdict.
Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.
Research requirements before AI-assisted coding
Produce a structured research handoff identifying external components, facts, risks and unresolved questions before implementation.
Before you start
- A project requirements document in a supported common specification filename or an explicit requirements path.
- An AI coding-agent environment that can run the /brainblast research skill.
Steps
Run the /brainblast research skill against the requirements document.
Let Brainblast identify external components and research official documentation, package registries and other planned sources.
Review the generated research, including risks, assumptions, resolved questions and requirements corrections.
Use the resulting final report and report.json as the handoff for coding.
Expected outcome
A research report and machine-readable report.json containing component findings, severity-tagged risks, pre-coding decisions and requirements corrections.
Guide limitations
- The supplied documentation does not specify the full command options or output formats beyond the documented report artifacts.
- The research workflow depends on the documented coding-agent entry point and source-browsing process.
Add integration-risk enforcement to CI
Scan written code and gate the build when documented integration risks remain.
Before you start
- A repository containing the code to scan.
- The brainblast npm CLI available through the documented npx entry point.
- A preceding research report or report.json when enforcing identified research findings.
Steps
Run npx brainblast . against the repository.
Use CI mode for a non-interactive run where required.
Apply the dependency-free gate to the generated report.json and select the documented failure threshold or blocked-verdict behavior.
Review the resulting build decision and remaining severity-tagged risks.
Expected outcome
CI receives an exit-code decision based on unresolved critical or other selected-severity risks, or on a blocked verdict.
Guide limitations
- The evidence does not specify a CI provider, configuration syntax, exact command-line options or complete output format.
- The documented default auditor is static and offline; executed and differential verification are opt-in paths.
Sources & review
Source record details
- Source availability
- Active
- Sources checked
- 1 October 2026
- Original website title
- DSB-117/brainblast · Official homepage title · retained in its source language