Roadmap¶
Audience: Contributors, reviewers, and stakeholders asking "what shipped, what's next, and why is Epic 4 missing?" Outcome: After reading, you can answer "is this in scope this quarter?" without asking a human. Last verified: 2026-04-22
This is the in-repo backlog. Day-to-day sprint tracking lives in issues; this file captures the load-bearing multi-week arcs. If you are about to propose a new initiative, add a stub under Proposed and open a PR for review — a roadmap entry without a PR is aspiration.
1. Shipped in April 2026 (v0.3.0)¶
v0.3.0 bundles five epics completed between 2026-04-01 and 2026-04-22. Full change lists are in CHANGELOG.md under [0.3.0].
| Epic | Theme | Headline |
|---|---|---|
| 1 | Platform Foundation | Typed protocols, consolidated settings, schema-drift CI gate, golden-output regression, OTEL parent-child trace assertion, CI ruff + 80% coverage enforcement. |
| 2 | Observable Execution | Typed event wire format, live DAG animation, StepNode B2 redesign, 5-field drill-down, Playwright streaming PR gate (5×), reconnect-replay test, time-to-first-span SLO + p95 gate, nightly 50× reliability. |
| 3 | DevEx / Windows | scripts/setup-dev.ps1 one-command bootstrap, port-guard, workspace-test-runner, workflow-linter, Windows Unicode CLI fix. |
| 5 | Console UI Polish | ASCII StatusBadge, useHotkeys, dashboard filter, empty/error/404 states, skip-to-main link, focus ring audit, paper-theme contrast QA, BDagMini SVG thumbnail. |
| 6 | Evaluation & Data Depth | Additive Pydantic v2 evaluation contracts, tokens_30d live stat, GET /runs/{filename}/evaluation, dataset sample endpoints, Evaluations rubric accordion, Datasets 3-pane browser. |
Implementation plans for Epics 1 and 2 are preserved as history in docs/superpowers/plans/. Epics 3, 5, and 6 did not have plans at time of execution; retrospective plan docs exist at:
The Epic 4 question¶
There is no Epic 4. The epic numbering jumps from Epic 3 to Epic 5 intentionally — no story, plan, commit, branch, or changelog entry uses the label "Epic 4" in this repository. The number was allocated during planning but never authored.
This is a tombstone, not a gap: do not retroactively renumber 5/6 down to 4/5, and do not reclaim "Epic 4" for a future initiative. Future epics continue from Epic 7.
Verification: git log --all --grep="epic.4" -i returns zero matches as of 2026-04-22. A repo-wide grep of source + docs returns the same.
2. In flight (Sprint B — targeted 2026-04-29 → 2026-05-10)¶
Sprint B is stabilization, not new capability. These items unblock a clean v0.3.1 point release.
| Item | Owner | Target |
|---|---|---|
Unmask 35 mypy findings in agentic-v2-eval/ — the eval package currently runs mypy with exemptions; cut that list to zero. |
unassigned | Sprint B |
Fix SLO p95 empty-window trivial-pass — time-to-first-span gate should require ≥ N samples before it declares "pass". See KNOWN_LIMITATIONS.md. |
unassigned | Sprint B |
Automate Python ↔ TypeScript wire-format drift detection — today the contracts/events.py union and ui/src/api/types.ts are mirrored by hand. Add a generator or diff test. |
unassigned | Sprint B |
| Dataset sample endpoint API polish — revisit query-param-for-slash-handling after we see real usage; possibly move to path-param with escaping. | unassigned | Sprint B |
Placeholder / no-LLM CI mode — currently CI depends on GITHUB_TOKEN + GitHub Models; a no-network mode would allow contributions from contributors without a token. |
unassigned | Sprint B or later |
None of these block v0.3.0 release; all are honest accounting of debt taken on to ship.
3. Proposed (Epic 7+)¶
Epic 7 — First-Run Experience¶
Problem statement: A contributor landing on this repo cannot reliably go from git clone to "I understand this and ran a workflow" in under 30 minutes without reading source code. The documentation work bundled in this docs sprint closes a significant portion of that gap, but does not ship the polish:
- Self-contained no-LLM demo path (
test_deterministicexists but is not surfaced as the recommended first step). - Dashboard "Hello World" tile on first run — show the user something without requiring them to craft a JSON input file.
- Better error messages when no provider key is set (current behavior: cryptic router failure deep in a call stack).
- Devcontainer validation in CI (exists but not yet a PR gate).
Definition of done:
- A contributor with no prior context completes
docs/ONBOARDING.mdQuick Start in < 10 minutes on a fresh Windows or Ubuntu clone. agentic run test_deterministicsucceeds with no environment variables set, producing clear output explaining the zero-LLM path.- The dashboard ships with a default guide panel that walks through the first run in < 3 clicks.
- Devcontainer builds on every PR touching
.devcontainer/.
Target: Not yet scheduled. Sprint B first.
Epic 8 — Production Readiness Pack (candidate)¶
Problem statement: This platform claims enterprise-grade practices for cleared federal environments. Remaining gaps before that claim holds end-to-end: authentication on the API server, tenant isolation, an audit log that is actually immutable (current logs are structured but file-based), and a supply-chain story for model weights.
Status: Candidate only. Needs scoping before it becomes an epic.
Epic 9 — Multi-Run Comparison (candidate)¶
Problem statement: Today the UI shows one run at a time. A frequent workflow during evaluation is comparing two or three runs side by side — same workflow, different prompts or adapters — to spot regressions. Some primitives exist (agentic compare, ADR-012 UI Evaluation Hub proposed), but the full comparison UX has not been built.
Status: Partially proposed in ADR-012. Consolidate into an epic if prioritized.
4. Out of scope for v0.3.x¶
Flagged here so nobody quietly adds them to a sprint:
- Presentation / deck system — extracted to a separate repo at
c:\Users\tandf\source\presenton 2026-04-22. SeeMIGRATIONS.md. Not returning to this repo. - Cross-language agent workers — this runtime is Python. Polyglot agents are intentionally deferred.
- Multi-tenant billing / quotas — scope for Epic 8 candidate only if it is promoted.
5. How to propose new work¶
- Open a PR that adds a subsection under §3 (Proposed). Include: problem statement, definition of done, rough sizing (story-weeks), open questions.
- If the work crosses an architectural boundary — new engine, new contract, new security surface — write an ADR under
adr/at the same time (seeCONTRIBUTING.md). - Once an epic is accepted for sprint, move its entry to §2 (In flight) and link the first sprint's plan doc under
superpowers/plans/. - When the epic ships, move its entry to §1 with the release it landed in.
An accepted roadmap entry is a promise to a reviewer, not to a user — shipping dates slip, scope narrows. Update the entry when reality changes.