This is the record of an abcd-managed repository

abcdBetav0.16.2 · 2026-11-06

Development record

itd-6 keeps a fixed RP MCP → Codex CLI → in-session cascade; the dependencies page says no such fixed cascade exists and that RepoPrompt, codex and a direct API oracle are opt-in adapters

issuewontfixmajoriss-2610262032144104

itd-6 keeps a fixed RP MCP → Codex CLI → in-session cascade; the dependencies page says no such fixed cascade exists and that RepoPrompt, codex and a direct API oracle are opt-in adapters

A premise contradiction (severity major) found by the cross-document consistency pass.

  • End A: .abcd/development/intents/shipped/itd-6-rp-mcp-only-integration.md:55 — "The cascade defined by this intent (RP MCP → Codex CLI → in-session subagent) is and stays availability-driven"
  • End B: .abcd/development/brief/02-constraints/02-dependencies.md:34 — "RepoPrompt and codex are two such adapters; a direct API oracle is a legitimate adapter too. There is no fixed RP→codex→in-session cascade."

itd-6, a shipped intent, says the cascade it defines (RP MCP → Codex CLI → in-session subagent) 'is and stays availability-driven'. 02-dependencies says RepoPrompt and codex are two opt-in adapters, that a direct API oracle is a legitimate adapter, and that there is no fixed RP→codex→in-session cascade. The oracle architecture cannot both keep that cascade and have none, and a builder following itd-6's scope would build the routing that the constraints chapter says does not exist.

Evidence: .abcd/work/reviews/2026-10-26-consistency/00-summary.md, finding 240.

Grounds

  • declined: Closed in the 2026-10-27 triage (dcn-2610271112330536): a later record already settles which end holds. The quoted cascade sentence sits in itd-6's original scope, which the record's own re-file note says is dropped and to be read through the re-file and Decisions; no fixed cascade exists in internal/core/oracle or the adapters.

Frontmatter

idiss-2610262032144104
typeissue
lifecyclewontfix
severitymajor

References1

mentions1