v6 readiness bar

[kubestellar-hive]

Strategic Finding

Type: roadmap-gap
Horizon: near-term

src/docs/v6-readiness.md (merged via #7675, resolving the gap #7674 named) says: "Create or maintain one issue named v6 readiness bar and keep its checklist in sync with the tables above." The v5 counterpart of that instruction sat unfulfilled until #6016 was retro-filed — this issue exists so v6 does not repeat that.

This issue is that tracker. Checklists below mirror src/docs/v6-readiness.md verbatim; rows are checked only with linked evidence (PR, workflow run, message screenshot/audit entry, pager event), never on "merged".

1. Channel binding

v6 earns its first moving tag only when:

• The v5 GA bar's Release-train rows are green (v5 GA bar #6016). v6 does not compete with v5 GA evidence for attention ([strategist] v6 opened before v5 GA: no stated priority between v6 tracks and the 4 remaining GA-bar evidence rows, with the v4 freeze backstop <4 weeks out #7577, ROADMAP.md).
• v6 is present in .github/release-lines.yml and the release-line guard is green on v6.
• Every shipped surface below has its guard-invariant conformance row and at least one live-exercise row checked with linked evidence.

2. Guard-invariant conformance (per surface)

Checked only with a link to a test that fails if the surface bypasses any of: dashboard role floor, mode ladder/capabilities at the proxy, Converse for replies, ioscan on inbound, canary/secret scrub on outbound.

• GitHub @-mention triggers (shipped ✨ mention: GitHub mention triggers phase 1 (poll-based) (#7483) #7582/✨ mention: phase 2 — in-thread replies when the summoned run completes (#7483) #7597/✨ mention: phase 3 — webhook accelerator for lower-latency summons (#7483) #7623)
• Slack Socket Mode (✨ slack: Socket Mode backend on the chat spine (#7563) #7585)
• Discord spine port + reliability (🌱 chat: extract the transport-agnostic chat spine from pkg/discord #7572/🐛 chat: reconnect and cancellation reliability parity across backends (#7563) #7586)
• Microsoft Teams (✨ msteams: chat backend on the pkg/chat spine (#7590) #7621)
• Matrix (✨ matrix: chat backend on the pkg/chat spine (#7591) #7617)
• Telegram (✨ telegram: chat backend on the pkg/chat spine (#7592) #7616)
• Email escalation, outbound + reply-to-act (✨ v6: pkg/escalate — severity-routed escalation fan-out with email + push sinks #7613/✨ escalate: severity-routed fan-out with email and push sinks (#7613) #7618)
• Push / on-call: ntfy / Pushover / PagerDuty (✨ v6: pkg/escalate — severity-routed escalation fan-out with email + push sinks #7613/✨ escalate: severity-routed fan-out with email and push sinks (#7613) #7618)

3. Live exercise (per surface)

One real round-trip against a live hive — not a unit test:

• GitHub @-mention: human mentions the App on a real issue/PR; kick runs; 👀 ack + audit entries linked.
• Slack: one command round-trip + one notification delivery over Socket Mode from a pull-only cluster.
• Discord: reconnect/backoff observed across one induced disconnect; notification parity spot-checked.
• Teams / Matrix / Telegram: one command round-trip + one notification delivery each.
• Email: one HUMAN DECISION NEEDED escalation delivered; one allowlisted inbound reply acted on (or reply-to-act explicitly deferred in the doc).
• Push / on-call: one requires_human verdict pages a real device via at least one provider.

4. Scope discipline

• New v6 tracks enter via the line epic (📝 v6 epic: dashboard-optional operation — meet humans where they are #7563); RFCs ([RFC] Archetypes for long-running tasks: name Oracle/Generator/Executor/Gate onto the staged convergence layer #7620, [RFC] [v6] Standby contributors: a lane paused for budget hands its queue to volunteers, behind a model floor #7629) are measured against this bar; work adding surfaces adds its conformance + exercise rows in the same PR.

Rationale

Every named v6 track merged within ~24h of line-open, but merged is not exercised: v6 publishes no channel and no surface has run against a live hive. The bar doc is only useful if its live tracker exists — this is where evidence accumulates. Mirrors the #6016 pattern for v5.

Proposed Next Step

Maintainers adopt this issue as the canonical v6 readiness bar tracker, link it from src/docs/v6-readiness.md's "Tracker checklist" section, and begin attaching conformance-test evidence for the surfaces that already have test suites (Discord 96.6% / chat 94.0% coverage per #7563 — a conformance-shaped test may be a small delta).

---

Filed by strategist agent (ACMM L5 — hold-gated mode)

— hive: agent=strategist backend=copilot model=claude-sonnet-4-6

[kubestellar-hive]

task-list sweep: 0 of 18 items ticked. Not closing yet — outstanding boxes remain.

Outstanding items:

• 🔲 The v5 GA bar's Release-train rows are green (v5 GA bar #6016). v6 does not compete with v5 GA evidence for attention ([strategist] v6 opened before v5 GA: no stated priority between v6 tracks and the 4 remaining GA-bar evidence rows, with the v4 freeze backstop <4 weeks out #7577, ROADMAP.md).
• 🔲 v6 is present in .github/release-lines.yml and the release-line guard is green on v6.
• 🔲 Every shipped surface below has its guard-invariant conformance row and at least one live-exercise row checked with linked evidence.
• 🔲 GitHub @-mention triggers (shipped ✨ mention: GitHub mention triggers phase 1 (poll-based) (#7483) #7582/✨ mention: phase 2 — in-thread replies when the summoned run completes (#7483) #7597/✨ mention: phase 3 — webhook accelerator for lower-latency summons (#7483) #7623)
• 🔲 Slack Socket Mode (✨ slack: Socket Mode backend on the chat spine (#7563) #7585)
• 🔲 Discord spine port + reliability (🌱 chat: extract the transport-agnostic chat spine from pkg/discord #7572/🐛 chat: reconnect and cancellation reliability parity across backends (#7563) #7586)
• 🔲 Microsoft Teams (✨ msteams: chat backend on the pkg/chat spine (#7590) #7621)
• 🔲 Matrix (✨ matrix: chat backend on the pkg/chat spine (#7591) #7617)
• 🔲 Telegram (✨ telegram: chat backend on the pkg/chat spine (#7592) #7616)
• 🔲 Email escalation, outbound + reply-to-act (✨ v6: pkg/escalate — severity-routed escalation fan-out with email + push sinks #7613/✨ escalate: severity-routed fan-out with email and push sinks (#7613) #7618)
• 🔲 Push / on-call: ntfy / Pushover / PagerDuty (✨ v6: pkg/escalate — severity-routed escalation fan-out with email + push sinks #7613/✨ escalate: severity-routed fan-out with email and push sinks (#7613) #7618)
• 🔲 GitHub @-mention: human mentions the App on a real issue/PR; kick runs; 👀 ack + audit entries linked.
• 🔲 Slack: one command round-trip + one notification delivery over Socket Mode from a pull-only cluster.
• 🔲 Discord: reconnect/backoff observed across one induced disconnect; notification parity spot-checked.
• 🔲 Teams / Matrix / Telegram: one command round-trip + one notification delivery each.
• 🔲 Email: one HUMAN DECISION NEEDED escalation delivered; one allowlisted inbound reply acted on (or reply-to-act explicitly deferred in the doc).
• 🔲 Push / on-call: one requires_human verdict pages a real device via at least one provider.
• 🔲 New v6 tracks enter via the line epic (📝 v6 epic: dashboard-optional operation — meet humans where they are #7563); RFCs ([RFC] Archetypes for long-running tasks: name Oracle/Generator/Executor/Gate onto the staged convergence layer #7620, [RFC] [v6] Standby contributors: a lane paused for budget hands its queue to volunteers, behind a model floor #7629) are measured against this bar; work adding surfaces adds its conformance + exercise rows in the same PR.

Merged PRs referencing this issue so far:

• [strategist] planning: link live tracker #7683 from v6-readiness.md #7684 — [strategist] planning: link live tracker v6 readiness bar #7683 from v6-readiness.md

This comment is edited in place by the task-list sweep on every cycle; it is not duplicated.