You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
id: design: provider security model (from #5686) #5905
Follow-on design ticket from #5686 — its decomposition sketch names this as a separate design, and the parent's acceptance requires these tickets to be filed and linked. Filed in the 2026-11-04 gap pass (WS-F) so the design work exists and can be prioritised on its own merits; implementation timing is a separate decision, possibly months out.
What this design must settle
Impersonation, rogue providers, capability confusion, and the ceiling-composition audit for the provider model.
Context from the parent
The parent design fixes the single capability-provider model — declaration, discovery, negotiation, least-privilege grants — instantiated by ESP32/MQTT, terminal, Bevy, smartphone, and web providers. The parent issue carries the surrounding decisions this ticket must not re-decide; where this ticket's answer would change a parent decision, record that explicitly rather than silently forking it.
Decision record expected
Close this design with a recorded decision (on this issue or in a doc/ record linked from it): the chosen shape, the alternatives rejected and why, and the invariants implementation must hold. If the design splits further, file the splits and link them here rather than growing this issue.
Relations
Parent: #5686. Programme: epic #2056. Filed by WS-F (2026-11-04 ID design gap & expansion pass).
Field survey filed: #5931 (capability-provider research widening, per Drew's 2026-11-04 ruling) covers the current option space relevant to this issue in §6 — security-model inputs: MCP's OAuth-centric authorization vs ID's owner-controlled grants, CapTP facet attenuation, SPIFFE identity plumbing, per-layer fit/fight analysis. It is an inventory, not a decision — the section ends with the open questions this design needs to answer.
Expectation: handed off to the opencode execution lane; the laptop claimant launches the bound opencode session, which drives task 10307 to its terminal state.
Triage: #5905 is an in-scope, still-open design child of #5686. Its acceptance is specific (settle impersonation, rogue-provider, capability-confusion, and ceiling-composition invariants, with alternatives and decision record); related research inputs are already filed in #5931. No PR or superseding implementation is linked. Keep this issue open for design prioritization; prior queue attempts were handoffs that never launched, so no design work was done here.
Activity
developing-today commentedon Nov 5, 2026
CODE-1588
dezren39 commentedon Nov 5, 2026
Field survey filed: #5931 (capability-provider research widening, per Drew's 2026-11-04 ruling) covers the current option space relevant to this issue in §6 — security-model inputs: MCP's OAuth-centric authorization vs ID's owner-controlled grants, CapTP facet attenuation, SPIFFE identity plumbing, per-layer fit/fight analysis. It is an inventory, not a decision — the section ends with the open questions this design needs to answer.
dezren39 commentedon Nov 5, 2026
🤖 Agent note · handed off
dezren39 commentedon Nov 5, 2026
🤖 Agent note · handed off
developing-today commentedon Nov 5, 2026
🤖 Agent note · handed off
dezren39 commentedon Nov 5, 2026
Triage: #5905 is an in-scope, still-open design child of #5686. Its acceptance is specific (settle impersonation, rogue-provider, capability-confusion, and ceiling-composition invariants, with alternatives and decision record); related research inputs are already filed in #5931. No PR or superseding implementation is linked. Keep this issue open for design prioritization; prior queue attempts were handoffs that never launched, so no design work was done here.