fix(codex): accept Codex 0.150 account plans

[rynfar]

Codex 0.150 reports six account plan types Pylon does not know:
self_serve_business_prolite, ent26, enterprise_cbp_automation,
edu_plus, edu_pro, and the prolite generator input. Adds them to the
PlanType definition in the schema generator and the generated schema, and maps
each to the right subscription label.

The two halves are coupled and must land together. Pylon's
codexAccountAuthLabel ends in account.planType satisfies never, so widening
the schema without widening the switch breaks the build. Verified rather than
assumed — see the mutation probe below.

Adopted from T3 Code 94401d01b956828eaa989ff4a80046c20d7b6088
(pingdotgg/t3code#8447), cherry-picked clean. Pylon's copy of this function was
byte-identical to upstream's, so no adaptation was needed even though
CodexProvider.ts carries ~111 lines of Pylon-only usage-limit work elsewhere.

Verification

• vp test run packages/effect-codex-app-server/src/schema.test.ts — 2 passed
• vp run -F t3 typecheck — clean (only pre-existing suggestion diagnostics
  in untouched files)
• Mutation probe: deleting just case "edu_plus": yields
  CodexProvider.ts(129,24): error TS1360: Type '"edu_plus"' does not satisfy the expected type 'never', confirming the schema/label coupling
• vp lint + vp fmt — the one Schema.is this adds is hoisted to module
  scope per no-inline-schema-compile; the 5 remaining warnings are
  pre-existing inline calls whose line numbers shifted

Note for maintainers, not fixed here

schema.gen.ts cannot be verified by regenerating it. UPSTREAM_REF is pinned
to 678157aca, the same ref in the file's header, yet node scripts/generate.ts
now produces ~39k lines of difference — a dependency emits
.annotate({ identifier: … }) and orders declarations differently than when the
file was last generated. This is pre-existing and shared with upstream: the
identifier: " count is 0 in the committed file, on origin/pylon, and on
t3code-upstream/main alike. Out of scope for this PR, but it means the
generated file is effectively hand-maintained today.

Claude Opus 5 via Claude Code.

---

^{Need help on this PR? Tag @codesmith-bot with what you need. Autofix is disabled.}

[github-actions]

Thread transfer impact

✅ Thread transfer remains within every enforced ceiling.

Provider	Metric	Main baseline	This PR	Impact	PR ceiling
Codex	Total thread wire	13.3 KiB	13.3 KiB	+67 B (+0.5%)	15.1 KiB	✅
Codex	Thread snapshot wire	6.9 KiB	6.9 KiB	+5 B (+0.1%)	7.3 KiB	✅
Codex	Live turn WebSocket wire	6.4 KiB	6.4 KiB	+62 B (+1.0%)	7.8 KiB	✅
Codex	Live turn WebSocket decoded	55.5 KiB	55.6 KiB	+44 B (+0.1%)	66.4 KiB	✅
Codex	Live turn messages	9	10	+1 (+11.1%)	21	✅
Claude	Total thread wire	13.3 KiB	13.3 KiB	+2 B (+0.0%)	15.1 KiB	✅
Claude	Thread snapshot wire	6.9 KiB	6.9 KiB	+21 B (+0.3%)	7.3 KiB	✅
Claude	Live turn WebSocket wire	6.5 KiB	6.4 KiB	−19 B (−0.3%)	7.8 KiB	✅
Claude	Live turn WebSocket decoded	56.4 KiB	56.4 KiB	−44 B (−0.1%)	66.4 KiB	✅
Claude	Live turn messages	11	10	−1 (−9.1%)	21	✅

Baseline: 3d6035f · PR result: 2c7ec99 · Source CI: success

Scenario and decoded snapshot size

10 historical turns, 5 command tools per turn, 878.9 KiB retained MCP result per historical turn, and a 1.05 MiB retained result in the measured turn.

• Codex decoded thread snapshot: 109.4 KiB
• Claude decoded thread snapshot: 110.1 KiB

Updated in place by a trusted workflow. PR artifacts are strictly validated and never executed.

[rynfar]

Codex 0.150 reports six account plan types Pylon does not know:
self_serve_business_prolite, ent26, enterprise_cbp_automation,
edu_plus, edu_pro, and the prolite generator input. Adds them to the
PlanType definition in the schema generator and the generated schema, and maps
each to the right subscription label.

The two halves are coupled and must land together. Pylon's
codexAccountAuthLabel ends in account.planType satisfies never, so widening
the schema without widening the switch breaks the build. Verified rather than
assumed — see the mutation probe below.

Adopted from T3 Code 94401d01b956828eaa989ff4a80046c20d7b6088
(pingdotgg/t3code#8447), cherry-picked clean. Pylon's copy of this function was
byte-identical to upstream's, so no adaptation was needed even though
CodexProvider.ts carries ~111 lines of Pylon-only usage-limit work elsewhere.

Verification

• vp test run packages/effect-codex-app-server/src/schema.test.ts — 2 passed
• vp run -F t3 typecheck — clean (only pre-existing suggestion diagnostics
  in untouched files)
• Mutation probe: deleting just case "edu_plus": yields
  CodexProvider.ts(129,24): error TS1360: Type '"edu_plus"' does not satisfy the expected type 'never', confirming the schema/label coupling
• vp lint + vp fmt — the one Schema.is this adds is hoisted to module
  scope per no-inline-schema-compile; the 5 remaining warnings are
  pre-existing inline calls whose line numbers shifted

Note for maintainers, not fixed here

schema.gen.ts cannot be verified by regenerating it. UPSTREAM_REF is pinned
to 678157aca, the same ref in the file's header, yet node scripts/generate.ts
now produces ~39k lines of difference — a dependency emits
.annotate({ identifier: … }) and orders declarations differently than when the
file was last generated. This is pre-existing and shared with upstream: the
identifier: " count is 0 in the committed file, on origin/pylon, and on
t3code-upstream/main alike. Out of scope for this PR, but it means the
generated file is effectively hand-maintained today.

Claude Opus 5 via Claude Code.

---

^{Need help on this PR? Tag @codesmith-bot with what you need. Autofix is disabled.}

[github-actions]

Thread transfer impact

✅ Thread transfer remains within every enforced ceiling.

Provider	Metric	Main baseline	This PR	Impact	PR ceiling
Codex	Total thread wire	13.3 KiB	13.3 KiB	+67 B (+0.5%)	15.1 KiB	✅
Codex	Thread snapshot wire	6.9 KiB	6.9 KiB	+5 B (+0.1%)	7.3 KiB	✅
Codex	Live turn WebSocket wire	6.4 KiB	6.4 KiB	+62 B (+1.0%)	7.8 KiB	✅
Codex	Live turn WebSocket decoded	55.5 KiB	55.6 KiB	+44 B (+0.1%)	66.4 KiB	✅
Codex	Live turn messages	9	10	+1 (+11.1%)	21	✅
Claude	Total thread wire	13.3 KiB	13.3 KiB	+2 B (+0.0%)	15.1 KiB	✅
Claude	Thread snapshot wire	6.9 KiB	6.9 KiB	+21 B (+0.3%)	7.3 KiB	✅
Claude	Live turn WebSocket wire	6.5 KiB	6.4 KiB	−19 B (−0.3%)	7.8 KiB	✅
Claude	Live turn WebSocket decoded	56.4 KiB	56.4 KiB	−44 B (−0.1%)	66.4 KiB	✅
Claude	Live turn messages	11	10	−1 (−9.1%)	21	✅

Baseline: 3d6035f · PR result: 2c7ec99 · Source CI: success

Scenario and decoded snapshot size

10 historical turns, 5 command tools per turn, 878.9 KiB retained MCP result per historical turn, and a 1.05 MiB retained result in the measured turn.

• Codex decoded thread snapshot: 109.4 KiB
• Claude decoded thread snapshot: 110.1 KiB

Updated in place by a trusted workflow. PR artifacts are strictly validated and never executed.

[rynfar]

Codex 0.150 reports six account plan types Pylon does not know:
self_serve_business_prolite, ent26, enterprise_cbp_automation,
edu_plus, edu_pro, and the prolite generator input. Adds them to the
PlanType definition in the schema generator and the generated schema, and maps
each to the right subscription label.

The two halves are coupled and must land together. Pylon's
codexAccountAuthLabel ends in account.planType satisfies never, so widening
the schema without widening the switch breaks the build. Verified rather than
assumed — see the mutation probe below.

Adopted from T3 Code 94401d01b956828eaa989ff4a80046c20d7b6088
(pingdotgg/t3code#8447), cherry-picked clean. Pylon's copy of this function was
byte-identical to upstream's, so no adaptation was needed even though
CodexProvider.ts carries ~111 lines of Pylon-only usage-limit work elsewhere.

Verification

• vp test run packages/effect-codex-app-server/src/schema.test.ts — 2 passed
• vp run -F t3 typecheck — clean (only pre-existing suggestion diagnostics
  in untouched files)
• Mutation probe: deleting just case "edu_plus": yields
  CodexProvider.ts(129,24): error TS1360: Type '"edu_plus"' does not satisfy the expected type 'never', confirming the schema/label coupling
• vp lint + vp fmt — the one Schema.is this adds is hoisted to module
  scope per no-inline-schema-compile; the 5 remaining warnings are
  pre-existing inline calls whose line numbers shifted

Note for maintainers, not fixed here

schema.gen.ts cannot be verified by regenerating it. UPSTREAM_REF is pinned
to 678157aca, the same ref in the file's header, yet node scripts/generate.ts
now produces ~39k lines of difference — a dependency emits
.annotate({ identifier: … }) and orders declarations differently than when the
file was last generated. This is pre-existing and shared with upstream: the
identifier: " count is 0 in the committed file, on origin/pylon, and on
t3code-upstream/main alike. Out of scope for this PR, but it means the
generated file is effectively hand-maintained today.

Claude Opus 5 via Claude Code.

---

^{Need help on this PR? Tag @codesmith-bot with what you need. Autofix is disabled.}

[github-actions]

Thread transfer impact

✅ Thread transfer remains within every enforced ceiling.

Provider	Metric	Main baseline	This PR	Impact	PR ceiling
Codex	Total thread wire	13.3 KiB	13.3 KiB	+67 B (+0.5%)	15.1 KiB	✅
Codex	Thread snapshot wire	6.9 KiB	6.9 KiB	+5 B (+0.1%)	7.3 KiB	✅
Codex	Live turn WebSocket wire	6.4 KiB	6.4 KiB	+62 B (+1.0%)	7.8 KiB	✅
Codex	Live turn WebSocket decoded	55.5 KiB	55.6 KiB	+44 B (+0.1%)	66.4 KiB	✅
Codex	Live turn messages	9	10	+1 (+11.1%)	21	✅
Claude	Total thread wire	13.3 KiB	13.3 KiB	+2 B (+0.0%)	15.1 KiB	✅
Claude	Thread snapshot wire	6.9 KiB	6.9 KiB	+21 B (+0.3%)	7.3 KiB	✅
Claude	Live turn WebSocket wire	6.5 KiB	6.4 KiB	−19 B (−0.3%)	7.8 KiB	✅
Claude	Live turn WebSocket decoded	56.4 KiB	56.4 KiB	−44 B (−0.1%)	66.4 KiB	✅
Claude	Live turn messages	11	10	−1 (−9.1%)	21	✅

Baseline: 3d6035f · PR result: 2c7ec99 · Source CI: success

Scenario and decoded snapshot size

10 historical turns, 5 command tools per turn, 878.9 KiB retained MCP result per historical turn, and a 1.05 MiB retained result in the measured turn.

• Codex decoded thread snapshot: 109.4 KiB
• Claude decoded thread snapshot: 110.1 KiB

Updated in place by a trusted workflow. PR artifacts are strictly validated and never executed.

[rynfar]

Codex 0.150 reports six account plan types Pylon does not know:
self_serve_business_prolite, ent26, enterprise_cbp_automation,
edu_plus, edu_pro, and the prolite generator input. Adds them to the
PlanType definition in the schema generator and the generated schema, and maps
each to the right subscription label.

The two halves are coupled and must land together. Pylon's
codexAccountAuthLabel ends in account.planType satisfies never, so widening
the schema without widening the switch breaks the build. Verified rather than
assumed — see the mutation probe below.

Adopted from T3 Code 94401d01b956828eaa989ff4a80046c20d7b6088
(pingdotgg/t3code#8447), cherry-picked clean. Pylon's copy of this function was
byte-identical to upstream's, so no adaptation was needed even though
CodexProvider.ts carries ~111 lines of Pylon-only usage-limit work elsewhere.

Verification

• vp test run packages/effect-codex-app-server/src/schema.test.ts — 2 passed
• vp run -F t3 typecheck — clean (only pre-existing suggestion diagnostics
  in untouched files)
• Mutation probe: deleting just case "edu_plus": yields
  CodexProvider.ts(129,24): error TS1360: Type '"edu_plus"' does not satisfy the expected type 'never', confirming the schema/label coupling
• vp lint + vp fmt — the one Schema.is this adds is hoisted to module
  scope per no-inline-schema-compile; the 5 remaining warnings are
  pre-existing inline calls whose line numbers shifted

Note for maintainers, not fixed here

schema.gen.ts cannot be verified by regenerating it. UPSTREAM_REF is pinned
to 678157aca, the same ref in the file's header, yet node scripts/generate.ts
now produces ~39k lines of difference — a dependency emits
.annotate({ identifier: … }) and orders declarations differently than when the
file was last generated. This is pre-existing and shared with upstream: the
identifier: " count is 0 in the committed file, on origin/pylon, and on
t3code-upstream/main alike. Out of scope for this PR, but it means the
generated file is effectively hand-maintained today.

Claude Opus 5 via Claude Code.

---

^{Need help on this PR? Tag @codesmith-bot with what you need. Autofix is disabled.}

[github-actions]

Thread transfer impact

✅ Thread transfer remains within every enforced ceiling.

Provider	Metric	Main baseline	This PR	Impact	PR ceiling
Codex	Total thread wire	13.3 KiB	13.3 KiB	+67 B (+0.5%)	15.1 KiB	✅
Codex	Thread snapshot wire	6.9 KiB	6.9 KiB	+5 B (+0.1%)	7.3 KiB	✅
Codex	Live turn WebSocket wire	6.4 KiB	6.4 KiB	+62 B (+1.0%)	7.8 KiB	✅
Codex	Live turn WebSocket decoded	55.5 KiB	55.6 KiB	+44 B (+0.1%)	66.4 KiB	✅
Codex	Live turn messages	9	10	+1 (+11.1%)	21	✅
Claude	Total thread wire	13.3 KiB	13.3 KiB	+2 B (+0.0%)	15.1 KiB	✅
Claude	Thread snapshot wire	6.9 KiB	6.9 KiB	+21 B (+0.3%)	7.3 KiB	✅
Claude	Live turn WebSocket wire	6.5 KiB	6.4 KiB	−19 B (−0.3%)	7.8 KiB	✅
Claude	Live turn WebSocket decoded	56.4 KiB	56.4 KiB	−44 B (−0.1%)	66.4 KiB	✅
Claude	Live turn messages	11	10	−1 (−9.1%)	21	✅

Baseline: 3d6035f · PR result: 2c7ec99 · Source CI: success

Scenario and decoded snapshot size

10 historical turns, 5 command tools per turn, 878.9 KiB retained MCP result per historical turn, and a 1.05 MiB retained result in the measured turn.

• Codex decoded thread snapshot: 109.4 KiB
• Claude decoded thread snapshot: 110.1 KiB

Updated in place by a trusted workflow. PR artifacts are strictly validated and never executed.

[rynfar]

Codex 0.150 reports six account plan types Pylon does not know:
self_serve_business_prolite, ent26, enterprise_cbp_automation,
edu_plus, edu_pro, and the prolite generator input. Adds them to the
PlanType definition in the schema generator and the generated schema, and maps
each to the right subscription label.

The two halves are coupled and must land together. Pylon's
codexAccountAuthLabel ends in account.planType satisfies never, so widening
the schema without widening the switch breaks the build. Verified rather than
assumed — see the mutation probe below.

Adopted from T3 Code 94401d01b956828eaa989ff4a80046c20d7b6088
(pingdotgg/t3code#8447), cherry-picked clean. Pylon's copy of this function was
byte-identical to upstream's, so no adaptation was needed even though
CodexProvider.ts carries ~111 lines of Pylon-only usage-limit work elsewhere.

Verification

• vp test run packages/effect-codex-app-server/src/schema.test.ts — 2 passed
• vp run -F t3 typecheck — clean (only pre-existing suggestion diagnostics
  in untouched files)
• Mutation probe: deleting just case "edu_plus": yields
  CodexProvider.ts(129,24): error TS1360: Type '"edu_plus"' does not satisfy the expected type 'never', confirming the schema/label coupling
• vp lint + vp fmt — the one Schema.is this adds is hoisted to module
  scope per no-inline-schema-compile; the 5 remaining warnings are
  pre-existing inline calls whose line numbers shifted

Note for maintainers, not fixed here

schema.gen.ts cannot be verified by regenerating it. UPSTREAM_REF is pinned
to 678157aca, the same ref in the file's header, yet node scripts/generate.ts
now produces ~39k lines of difference — a dependency emits
.annotate({ identifier: … }) and orders declarations differently than when the
file was last generated. This is pre-existing and shared with upstream: the
identifier: " count is 0 in the committed file, on origin/pylon, and on
t3code-upstream/main alike. Out of scope for this PR, but it means the
generated file is effectively hand-maintained today.

Claude Opus 5 via Claude Code.

---

^{Need help on this PR? Tag @codesmith-bot with what you need. Autofix is disabled.}

[github-actions]

Thread transfer impact

✅ Thread transfer remains within every enforced ceiling.

Provider	Metric	Main baseline	This PR	Impact	PR ceiling
Codex	Total thread wire	13.3 KiB	13.3 KiB	+67 B (+0.5%)	15.1 KiB	✅
Codex	Thread snapshot wire	6.9 KiB	6.9 KiB	+5 B (+0.1%)	7.3 KiB	✅
Codex	Live turn WebSocket wire	6.4 KiB	6.4 KiB	+62 B (+1.0%)	7.8 KiB	✅
Codex	Live turn WebSocket decoded	55.5 KiB	55.6 KiB	+44 B (+0.1%)	66.4 KiB	✅
Codex	Live turn messages	9	10	+1 (+11.1%)	21	✅
Claude	Total thread wire	13.3 KiB	13.3 KiB	+2 B (+0.0%)	15.1 KiB	✅
Claude	Thread snapshot wire	6.9 KiB	6.9 KiB	+21 B (+0.3%)	7.3 KiB	✅
Claude	Live turn WebSocket wire	6.5 KiB	6.4 KiB	−19 B (−0.3%)	7.8 KiB	✅
Claude	Live turn WebSocket decoded	56.4 KiB	56.4 KiB	−44 B (−0.1%)	66.4 KiB	✅
Claude	Live turn messages	11	10	−1 (−9.1%)	21	✅

Baseline: 3d6035f · PR result: 2c7ec99 · Source CI: success

Scenario and decoded snapshot size

10 historical turns, 5 command tools per turn, 878.9 KiB retained MCP result per historical turn, and a 1.05 MiB retained result in the measured turn.

• Codex decoded thread snapshot: 109.4 KiB
• Claude decoded thread snapshot: 110.1 KiB

Updated in place by a trusted workflow. PR artifacts are strictly validated and never executed.

[rynfar]

Codex 0.150 reports six account plan types Pylon does not know:
self_serve_business_prolite, ent26, enterprise_cbp_automation,
edu_plus, edu_pro, and the prolite generator input. Adds them to the
PlanType definition in the schema generator and the generated schema, and maps
each to the right subscription label.

The two halves are coupled and must land together. Pylon's
codexAccountAuthLabel ends in account.planType satisfies never, so widening
the schema without widening the switch breaks the build. Verified rather than
assumed — see the mutation probe below.

Adopted from T3 Code 94401d01b956828eaa989ff4a80046c20d7b6088
(pingdotgg/t3code#8447), cherry-picked clean. Pylon's copy of this function was
byte-identical to upstream's, so no adaptation was needed even though
CodexProvider.ts carries ~111 lines of Pylon-only usage-limit work elsewhere.

Verification

• vp test run packages/effect-codex-app-server/src/schema.test.ts — 2 passed
• vp run -F t3 typecheck — clean (only pre-existing suggestion diagnostics
  in untouched files)
• Mutation probe: deleting just case "edu_plus": yields
  CodexProvider.ts(129,24): error TS1360: Type '"edu_plus"' does not satisfy the expected type 'never', confirming the schema/label coupling
• vp lint + vp fmt — the one Schema.is this adds is hoisted to module
  scope per no-inline-schema-compile; the 5 remaining warnings are
  pre-existing inline calls whose line numbers shifted

Note for maintainers, not fixed here

schema.gen.ts cannot be verified by regenerating it. UPSTREAM_REF is pinned
to 678157aca, the same ref in the file's header, yet node scripts/generate.ts
now produces ~39k lines of difference — a dependency emits
.annotate({ identifier: … }) and orders declarations differently than when the
file was last generated. This is pre-existing and shared with upstream: the
identifier: " count is 0 in the committed file, on origin/pylon, and on
t3code-upstream/main alike. Out of scope for this PR, but it means the
generated file is effectively hand-maintained today.

Claude Opus 5 via Claude Code.

---

^{Need help on this PR? Tag @codesmith-bot with what you need. Autofix is disabled.}

[github-actions]

Thread transfer impact

✅ Thread transfer remains within every enforced ceiling.

Provider	Metric	Main baseline	This PR	Impact	PR ceiling
Codex	Total thread wire	13.3 KiB	13.3 KiB	+67 B (+0.5%)	15.1 KiB	✅
Codex	Thread snapshot wire	6.9 KiB	6.9 KiB	+5 B (+0.1%)	7.3 KiB	✅
Codex	Live turn WebSocket wire	6.4 KiB	6.4 KiB	+62 B (+1.0%)	7.8 KiB	✅
Codex	Live turn WebSocket decoded	55.5 KiB	55.6 KiB	+44 B (+0.1%)	66.4 KiB	✅
Codex	Live turn messages	9	10	+1 (+11.1%)	21	✅
Claude	Total thread wire	13.3 KiB	13.3 KiB	+2 B (+0.0%)	15.1 KiB	✅
Claude	Thread snapshot wire	6.9 KiB	6.9 KiB	+21 B (+0.3%)	7.3 KiB	✅
Claude	Live turn WebSocket wire	6.5 KiB	6.4 KiB	−19 B (−0.3%)	7.8 KiB	✅
Claude	Live turn WebSocket decoded	56.4 KiB	56.4 KiB	−44 B (−0.1%)	66.4 KiB	✅
Claude	Live turn messages	11	10	−1 (−9.1%)	21	✅

Baseline: 3d6035f · PR result: 2c7ec99 · Source CI: success

Scenario and decoded snapshot size

10 historical turns, 5 command tools per turn, 878.9 KiB retained MCP result per historical turn, and a 1.05 MiB retained result in the measured turn.

• Codex decoded thread snapshot: 109.4 KiB
• Claude decoded thread snapshot: 110.1 KiB

Updated in place by a trusted workflow. PR artifacts are strictly validated and never executed.

[rynfar]

Codex 0.150 reports six account plan types Pylon does not know:
self_serve_business_prolite, ent26, enterprise_cbp_automation,
edu_plus, edu_pro, and the prolite generator input. Adds them to the
PlanType definition in the schema generator and the generated schema, and maps
each to the right subscription label.

The two halves are coupled and must land together. Pylon's
codexAccountAuthLabel ends in account.planType satisfies never, so widening
the schema without widening the switch breaks the build. Verified rather than
assumed — see the mutation probe below.

Adopted from T3 Code 94401d01b956828eaa989ff4a80046c20d7b6088
(pingdotgg/t3code#8447), cherry-picked clean. Pylon's copy of this function was
byte-identical to upstream's, so no adaptation was needed even though
CodexProvider.ts carries ~111 lines of Pylon-only usage-limit work elsewhere.

Verification

• vp test run packages/effect-codex-app-server/src/schema.test.ts — 2 passed
• vp run -F t3 typecheck — clean (only pre-existing suggestion diagnostics
  in untouched files)
• Mutation probe: deleting just case "edu_plus": yields
  CodexProvider.ts(129,24): error TS1360: Type '"edu_plus"' does not satisfy the expected type 'never', confirming the schema/label coupling
• vp lint + vp fmt — the one Schema.is this adds is hoisted to module
  scope per no-inline-schema-compile; the 5 remaining warnings are
  pre-existing inline calls whose line numbers shifted

Note for maintainers, not fixed here

schema.gen.ts cannot be verified by regenerating it. UPSTREAM_REF is pinned
to 678157aca, the same ref in the file's header, yet node scripts/generate.ts
now produces ~39k lines of difference — a dependency emits
.annotate({ identifier: … }) and orders declarations differently than when the
file was last generated. This is pre-existing and shared with upstream: the
identifier: " count is 0 in the committed file, on origin/pylon, and on
t3code-upstream/main alike. Out of scope for this PR, but it means the
generated file is effectively hand-maintained today.

Claude Opus 5 via Claude Code.

---

^{Need help on this PR? Tag @codesmith-bot with what you need. Autofix is disabled.}

[github-actions]

Thread transfer impact

✅ Thread transfer remains within every enforced ceiling.

Provider	Metric	Main baseline	This PR	Impact	PR ceiling
Codex	Total thread wire	13.3 KiB	13.3 KiB	+67 B (+0.5%)	15.1 KiB	✅
Codex	Thread snapshot wire	6.9 KiB	6.9 KiB	+5 B (+0.1%)	7.3 KiB	✅
Codex	Live turn WebSocket wire	6.4 KiB	6.4 KiB	+62 B (+1.0%)	7.8 KiB	✅
Codex	Live turn WebSocket decoded	55.5 KiB	55.6 KiB	+44 B (+0.1%)	66.4 KiB	✅
Codex	Live turn messages	9	10	+1 (+11.1%)	21	✅
Claude	Total thread wire	13.3 KiB	13.3 KiB	+2 B (+0.0%)	15.1 KiB	✅
Claude	Thread snapshot wire	6.9 KiB	6.9 KiB	+21 B (+0.3%)	7.3 KiB	✅
Claude	Live turn WebSocket wire	6.5 KiB	6.4 KiB	−19 B (−0.3%)	7.8 KiB	✅
Claude	Live turn WebSocket decoded	56.4 KiB	56.4 KiB	−44 B (−0.1%)	66.4 KiB	✅
Claude	Live turn messages	11	10	−1 (−9.1%)	21	✅

Baseline: 3d6035f · PR result: 2c7ec99 · Source CI: success

Scenario and decoded snapshot size

10 historical turns, 5 command tools per turn, 878.9 KiB retained MCP result per historical turn, and a 1.05 MiB retained result in the measured turn.

• Codex decoded thread snapshot: 109.4 KiB
• Claude decoded thread snapshot: 110.1 KiB

Updated in place by a trusted workflow. PR artifacts are strictly validated and never executed.

[rynfar]

Codex 0.150 reports six account plan types Pylon does not know:
self_serve_business_prolite, ent26, enterprise_cbp_automation,
edu_plus, edu_pro, and the prolite generator input. Adds them to the
PlanType definition in the schema generator and the generated schema, and maps
each to the right subscription label.

The two halves are coupled and must land together. Pylon's
codexAccountAuthLabel ends in account.planType satisfies never, so widening
the schema without widening the switch breaks the build. Verified rather than
assumed — see the mutation probe below.

Adopted from T3 Code 94401d01b956828eaa989ff4a80046c20d7b6088
(pingdotgg/t3code#8447), cherry-picked clean. Pylon's copy of this function was
byte-identical to upstream's, so no adaptation was needed even though
CodexProvider.ts carries ~111 lines of Pylon-only usage-limit work elsewhere.

Verification

• vp test run packages/effect-codex-app-server/src/schema.test.ts — 2 passed
• vp run -F t3 typecheck — clean (only pre-existing suggestion diagnostics
  in untouched files)
• Mutation probe: deleting just case "edu_plus": yields
  CodexProvider.ts(129,24): error TS1360: Type '"edu_plus"' does not satisfy the expected type 'never', confirming the schema/label coupling
• vp lint + vp fmt — the one Schema.is this adds is hoisted to module
  scope per no-inline-schema-compile; the 5 remaining warnings are
  pre-existing inline calls whose line numbers shifted

Note for maintainers, not fixed here

schema.gen.ts cannot be verified by regenerating it. UPSTREAM_REF is pinned
to 678157aca, the same ref in the file's header, yet node scripts/generate.ts
now produces ~39k lines of difference — a dependency emits
.annotate({ identifier: … }) and orders declarations differently than when the
file was last generated. This is pre-existing and shared with upstream: the
identifier: " count is 0 in the committed file, on origin/pylon, and on
t3code-upstream/main alike. Out of scope for this PR, but it means the
generated file is effectively hand-maintained today.

Claude Opus 5 via Claude Code.

---

^{Need help on this PR? Tag @codesmith-bot with what you need. Autofix is disabled.}

[github-actions]

Thread transfer impact

✅ Thread transfer remains within every enforced ceiling.

Provider	Metric	Main baseline	This PR	Impact	PR ceiling
Codex	Total thread wire	13.3 KiB	13.3 KiB	+67 B (+0.5%)	15.1 KiB	✅
Codex	Thread snapshot wire	6.9 KiB	6.9 KiB	+5 B (+0.1%)	7.3 KiB	✅
Codex	Live turn WebSocket wire	6.4 KiB	6.4 KiB	+62 B (+1.0%)	7.8 KiB	✅
Codex	Live turn WebSocket decoded	55.5 KiB	55.6 KiB	+44 B (+0.1%)	66.4 KiB	✅
Codex	Live turn messages	9	10	+1 (+11.1%)	21	✅
Claude	Total thread wire	13.3 KiB	13.3 KiB	+2 B (+0.0%)	15.1 KiB	✅
Claude	Thread snapshot wire	6.9 KiB	6.9 KiB	+21 B (+0.3%)	7.3 KiB	✅
Claude	Live turn WebSocket wire	6.5 KiB	6.4 KiB	−19 B (−0.3%)	7.8 KiB	✅
Claude	Live turn WebSocket decoded	56.4 KiB	56.4 KiB	−44 B (−0.1%)	66.4 KiB	✅
Claude	Live turn messages	11	10	−1 (−9.1%)	21	✅

Baseline: 3d6035f · PR result: 2c7ec99 · Source CI: success

Scenario and decoded snapshot size

10 historical turns, 5 command tools per turn, 878.9 KiB retained MCP result per historical turn, and a 1.05 MiB retained result in the measured turn.

• Codex decoded thread snapshot: 109.4 KiB
• Claude decoded thread snapshot: 110.1 KiB

Updated in place by a trusted workflow. PR artifacts are strictly validated and never executed.