This roadmap explains how the 16-repository enterprise product testbed can mature into credible AAIF/Linux Foundation-adjacent evidence.
It is not an AAIF proposal, not a Linux Foundation proposal, and not a claim of endorsement. It is a public operating plan for earning recognition through useful work, reviewable artifacts, and maintainer feedback.
Recognition should come from repeated evidence that helps the ecosystem:
- Serious repositories expose clear agentic boundaries.
- Paid-provider defaults have OSS and self-hosted alternatives.
- Developers can fork, run, inspect, and improve the work.
- Upstream communities receive small, useful issues and PRs.
- Claims stay behind evidence until maintainers merge or adopt work.
The portfolio is useful because it tests the profile across different enterprise surfaces instead of one demo app.
| AAIF-facing concern | Portfolio evidence surface | Primary repositories |
|---|---|---|
| Observability & Traceability | audit events, trace boundaries, provenance, reviewable evidence | SecureIDE Workbench, FlowGrid Orchestrator, SupportDesk Intelligence, Meeting Intelligence Hub |
| Identity & Trust | RBAC, tenant context, session policy, approval authority, access review | IdentityCore Platform, AccessGovernance Kit, TenantCommerce Control, SecureIDE Workbench |
| Security & Privacy | secret safety, sandboxing, PII boundaries, data retention, provider replacement | SecureIDE Workbench, SupportDesk Intelligence, LedgerOps Intelligence, KnowledgeStream Hub |
| Governance, Risk & Regulatory Alignment | human approval, policy posture, evidence retention, release gates | AccessGovernance Kit, LedgerOps Intelligence, WorkGrid Portfolio, SkillForge Learning |
| Workflows & Process Integration | connector governance, retries, escalation, handoffs, replay boundaries | FlowGrid Orchestrator, WorkGrid Portfolio, CollabBoard Command, Meeting Intelligence Hub |
| Accuracy & Reliability | answer provenance, validation evidence, regression checks, fallback behavior | SupportDesk Intelligence, SkillForge Learning, KnowledgeStream Hub, AppForge Studio |
| Agentic Commerce | tenant storefront control, catalog governance, entitlement boundaries | TenantCommerce Control, LedgerOps Intelligence, BrandOps Studio |
| Stage | Evidence required | Current posture |
|---|---|---|
| Portfolio hygiene | public repos, README onboarding, OSS alternatives, secret-safe setup, scoped roadmaps | In progress across the 16-repo testbed |
| Profile evidence | readiness JSON, boundary model, governance, security, validation notes | All 16 portfolio repositories now publish repo-local boundary evidence |
| Upstream usefulness | small AAIF/LF-adjacent PRs and issues that solve repository-local needs | Active contribution ledger tracks open work |
| External review | public discussion, focused review issue, maintainer review kit, review questions | Published but still awaiting independent feedback |
| External adoption | at least two non-Karuna maintainers try the profile or provide opt-in examples | Trial protocol published; external trial evidence still missing |
| Recognition candidate | merged upstream contributions, external reviewers, sponsor path, governance clarity | Not ready yet |
- Keep the contribution ledger current as AAIF and adjacent PRs receive review.
- Keep dependency-license inventory and release checklist current.
- Use the maintainer/reviewer growth policy when external reviewers engage.
- Keep all 16 portfolio repositories at profile level 3 evidence quality.
- Avoid new formal proposal language until external reviewers engage.
- Convert boundary evidence into lightweight test fixtures for the highest-risk workflow in each repository.
- Add one focused adoption issue asking external maintainers what evidence is missing.
- Publish one short case study per flagship repository, focused on boundaries and proof.
- Ask for correction in AAIF/LF-adjacent forums using public links, not cold tagging.
- Recruit two external reviewers or opt-in adopters.
- Convert accepted upstream feedback into profile changes.
- Add an explicit maintainer growth policy.
- Decide whether the work should remain independent, become example guidance, or prepare for a future formal proposal.
Do not claim recognition until the relevant gate is public:
| Claim | Required public evidence |
|---|---|
| Upstream contributor | PR merged and GitHub attributes the commit to @kmishra1204 |
| AAIF-adjacent reviewer | public maintainer comment, review, or issue response |
| External adoption | external repository, maintainer, or issue explicitly opts in |
| Foundation fit | AAIF/LF-adjacent maintainers say the shape is useful or worth routing |
| Formal proposal readiness | external adoption, governance clarity, sponsor path, and user decision on legal/account implications |
- AAIF Observability PR
#17: contributor onboarding guide. - AAIF Observability PR
#18: liaison roster and cross-WG intake tracker. - AAIF Taxonomy & Landscape PR
#23: term style guide. - AAIF Taxonomy & Landscape PR
#24: AI usage guide. - Public readiness profile releases through
v0.5.19. - Twelve month milestone plan published for public proposal-readiness review.
- All 16 portfolio repositories now publish repo-local
docs/AGENTIC_BOUNDARY_MODEL.mdevidence. - Public contribution ledger with recognition gates and anti-claim rules.
- AGENTS.md enterprise evidence checklist published as the first monthly ambassador-style educational artifact.
All listed AAIF PRs are work under review, not accepted contribution.
- Do not mass-tag maintainers.
- Do not claim AAIF or Linux Foundation affiliation.
- Do not submit a formal proposal before external review and adoption evidence exist.
- Do not treat the 16-repo portfolio as production proof by itself.
- Do not turn the readiness profile into marketing copy at the expense of runnable evidence.
The useful ask for reviewers is narrow:
Which evidence would make this profile useful enough for another agentic AI repository to try?
That question keeps the work grounded in ecosystem value instead of personal promotion.