docks-kit 0.22.25 → 0.23.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (77) hide show
  1. package/AGENTS.md +35 -81
  2. package/README.md +31 -58
  3. package/cli/docs/command-guards.md +231 -197
  4. package/cli/docs/flags.md +25 -55
  5. package/cli/docs/install.md +2 -3
  6. package/cli/docs/omp-context.md +5 -5
  7. package/cli/docs/omp-models.md +249 -48
  8. package/cli/docs/overview.md +4 -12
  9. package/cli/docs/platforms.md +2 -2
  10. package/cli/docs/sync-layers.md +10 -57
  11. package/cli/docs/toolchain.md +9 -78
  12. package/cli/src/argv.ts +0 -8
  13. package/cli/src/argvValidate.ts +2 -46
  14. package/cli/src/commands/docs.ts +2 -14
  15. package/cli/src/commands/harnesses.ts +6 -9
  16. package/cli/src/commands/sync.ts +5 -99
  17. package/cli/src/commands/toolchain.ts +4 -12
  18. package/cli/src/engine-native/DESIGN.md +67 -88
  19. package/cli/src/engine-native/deps.ts +6 -86
  20. package/cli/src/engine-native/engineCtx.ts +7 -76
  21. package/cli/src/engine-native/exec.ts +1 -18
  22. package/cli/src/engine-native/harnesses.ts +6 -6
  23. package/cli/src/engine-native/index.ts +4 -6
  24. package/cli/src/engine-native/jq.ts +4 -37
  25. package/cli/src/engine-native/modes.ts +3 -132
  26. package/cli/src/engine-native/ompRemovals.ts +14 -0
  27. package/cli/src/engine-native/os/darwin.ts +0 -4
  28. package/cli/src/engine-native/os/index.ts +1 -2
  29. package/cli/src/engine-native/os/linux.ts +0 -3
  30. package/cli/src/engine-native/os/posix.ts +2 -46
  31. package/cli/src/engine-native/os/types.ts +1 -32
  32. package/cli/src/engine-native/os/windows.ts +0 -49
  33. package/cli/src/engine-native/parseArgs.ts +11 -124
  34. package/cli/src/engine-native/parseHelp.ts +5 -40
  35. package/cli/src/engine-native/scratch.ts +1 -1
  36. package/cli/src/engine-native/services.ts +1 -2
  37. package/cli/src/engine-native/sharedTypes.d.ts +0 -40
  38. package/cli/src/engine-native/skillsSync.ts +1 -1
  39. package/cli/src/engine-native/syncConcurrency.ts +3 -15
  40. package/cli/src/engine-native/syncDispatch.ts +10 -57
  41. package/cli/src/engine-native/toolchain.ts +5 -40
  42. package/cli/src/engine.ts +0 -75
  43. package/cli/src/generated/sotPayload.ts +14 -41
  44. package/cli/src/main.ts +7 -19
  45. package/cli/src/manifests.ts +1 -93
  46. package/cli/src/payload.ts +2 -8
  47. package/cli/tsconfig.json +1 -3
  48. package/package.json +3 -5
  49. package/cli/docs/models.md +0 -58
  50. package/cli/docs/modifiers.md +0 -52
  51. package/cli/docs/plugins.md +0 -70
  52. package/cli/src/commands/model.ts +0 -68
  53. package/cli/src/commands/models.ts +0 -75
  54. package/cli/src/commands/plugins.ts +0 -35
  55. package/cli/src/commands/status.ts +0 -206
  56. package/cli/src/efforts.ts +0 -89
  57. package/cli/src/engine-native/claudeLsp.ts +0 -301
  58. package/cli/src/engine-native/claudeOptionalPlugins.ts +0 -126
  59. package/cli/src/engine-native/claudePluginPasses.ts +0 -426
  60. package/cli/src/engine-native/claudeRemovals.ts +0 -383
  61. package/cli/src/engine-native/claudeRetired.ts +0 -165
  62. package/cli/src/engine-native/claudeRuntime.ts +0 -157
  63. package/cli/src/engine-native/claudeSettings.ts +0 -201
  64. package/cli/src/engine-native/claudeSettingsModifiers.ts +0 -115
  65. package/cli/src/engine-native/claudeSync.ts +0 -312
  66. package/cli/src/engine-native/codexConfig.ts +0 -214
  67. package/cli/src/engine-native/codexHooks.ts +0 -91
  68. package/cli/src/engine-native/codexPaths.ts +0 -14
  69. package/cli/src/engine-native/codexPlugins.ts +0 -315
  70. package/cli/src/engine-native/codexStatus.ts +0 -25
  71. package/cli/src/engine-native/codexSync.ts +0 -314
  72. package/cli/src/engine-native/codexToml.ts +0 -303
  73. package/cli/src/engine-native/liveModels.ts +0 -264
  74. package/cli/src/engine-native/models.ts +0 -101
  75. package/cli/src/engine-native/parseModifiers.ts +0 -225
  76. package/cli/src/engine-native/settings.ts +0 -42
  77. package/cli/src/engine-native/tomlLines.ts +0 -92
package/cli/docs/flags.md CHANGED
@@ -1,62 +1,55 @@
1
1
  # Flag reference
2
2
 
3
- Convention: global flags are unprefixed; anything touching one tool's
3
+ Convention: global flags are unprefixed; a flag that touches one tool's
4
4
  deployed config is `--<tool>-<setting>`.
5
5
 
6
6
  ## Targets (positional)
7
7
 
8
8
  ```
9
9
  docks-kit sync # stored harness selection
10
- docks-kit sync claude # one
11
- docks-kit sync claude agents # two
12
- docks-kit sync omp # opt-in harness
10
+ docks-kit sync omp # one
11
+ docks-kit sync agents omp # two
13
12
  ```
14
13
 
15
- Positional targets are `claude`, `codex`, `agents`, and `omp`. A flag-less
16
- `sync` deploys the per-machine harness selection stored in
17
- `~/.docks-kit/kit.db`. A missing selection or an unreadable store selects `claude`,
18
- `codex`, and `agents`, and never selects `omp`. Choose the stored selection
19
- with `docks-kit harnesses`. With no stored selection, its picker pre-ticks all
20
- four harnesses, `omp` included; a flag-less `sync` does not change until the
21
- picker saves a selection. In the picker, Enter or Space ticks or unticks the
22
- harness under the cursor, Enter on the Save row stores the selection, and
23
- Escape leaves it unchanged.
14
+ Positional targets are `agents` and `omp`. A flag-less `sync` deploys the
15
+ per-machine harness selection stored in `~/.docks-kit/kit.db`. A missing
16
+ selection or an unreadable store selects `agents` and `omp`. Choose the stored
17
+ selection with `docks-kit harnesses`. With no stored selection, its picker
18
+ pre-ticks both harnesses; a flag-less `sync` does not change until the picker
19
+ saves a selection. In the picker, Enter or Space ticks or unticks the harness
20
+ under the cursor, Enter on the Save row stores the selection, and Escape
21
+ leaves it unchanged.
24
22
 
25
23
  ## Global flags
26
24
 
27
25
  | Flag | Effect |
28
26
  |------|--------|
29
27
  | `--dry-run` | Preview without applying; takes no sync lock and no folder lock |
30
- | `--reconcile` | Settings layer reconciled toward SoT (SoT keys win; user-only keys preserved; permissions arrays replaced) |
31
- | `--prune` | Uninstall kit-managed installs not in SoT: plugins, marketplaces, and the universal skills that `~/.agents/.kit-owned-skills` names |
32
- | `--skip-bubblewrap` | Skip optional bubblewrap bootstrap (Codex Linux sandbox) |
33
- | `--skip-plugin-refresh` | Install missing Claude, Codex, and omp plugins but skip refresh-only updates |
34
- | `--verbose` / `-v` | Also print no-op confirmations (already in sync, up to date, left as-is); accepted on `sync`, `model`, and `toolchain` |
28
+ | `--prune` | Remove the kit-managed universal skills that `~/.agents/.kit-owned-skills` names and SoT no longer lists |
29
+ | `--skip-plugin-refresh` | Install missing omp plugins but skip refresh-only updates |
30
+ | `--verbose` / `-v` | Also print no-op confirmations (already in sync, up to date, left as-is); accepted on `sync` and `toolchain` |
35
31
 
36
32
  ## Environment overrides
37
33
 
38
34
  | Variable | Effect |
39
35
  |----------|--------|
40
- | `DOCKS_KIT_SYNC_CONCURRENCY=1\|2\|3` | Maximum selected sync pipelines running at once. Default `3`; use `1` for serial golden/debug execution. Invalid values fail `sync` with exit 2 but do not affect `model` or `toolchain`. |
41
- | `DOCKS_KIT_SYNC_LOCK_WAIT_MS=<ms>` | Time that `sync`, a `model` write, and `omp --fast` wait for another sync to release a sync lock. A non-negative safe integer in milliseconds (at most `9007199254740991`); unset or empty means `120000` (120 s). Any other value prints `DOCKS_KIT_SYNC_LOCK_WAIT_MS must be a non-negative safe integer` and exits 2. |
36
+ | `DOCKS_KIT_SYNC_CONCURRENCY=1\|2\|3` | Maximum selected sync pipelines running at once. Default `3`; use `1` for serial golden/debug execution. Invalid values fail `sync` with exit 2 but do not affect `toolchain`. |
37
+ | `DOCKS_KIT_SYNC_LOCK_WAIT_MS=<ms>` | Time that `sync` and `omp --fast` wait for another sync to release a sync lock. A non-negative safe integer in milliseconds (at most `9007199254740991`); unset or empty means `120000` (120 s). Any other value prints `DOCKS_KIT_SYNC_LOCK_WAIT_MS must be a non-negative safe integer` and exits 2. |
42
38
 
43
39
  ## Sync lock
44
40
 
45
- `sync` and `model <tool> <value>` hold the lock
46
- `~/.cache/docks-kit/sync-lock.db` while they run, so two runs never write the
47
- same files at once.
41
+ `sync` holds the lock `~/.cache/docks-kit/sync-lock.db` while it runs, so two
42
+ runs never write the same files at once.
48
43
 
49
44
  - A run waits up to 120 s for another run to release the lock
50
45
  (`DOCKS_KIT_SYNC_LOCK_WAIT_MS` changes the wait). It then prints
51
46
  `another docks-kit sync holds <path>; run sync again after it ends` and
52
47
  exits 1 with no change.
53
- - Two homes can share a Codex folder (`CODEX_HOME`), an agents folder
54
- (`AGENTS_DIR`), or an omp folder (`PI_CODING_AGENT_DIR`, `PI_CONFIG_DIR`,
55
- `XDG_DATA_HOME`, or a symlink). So a writing sync also holds
56
- `.docks-kit-sync-lock.db` inside each such folder that it writes. For `codex`
57
- this is the Codex folder, and for `agents` the agents folder. For `omp` it is
58
- the omp agent folder, the intercom folder, and the omp data root. A
59
- `model codex <value>` write holds the Codex folder lock too. The held line
48
+ - Two homes can share an agents folder (`AGENTS_DIR`) or an omp folder
49
+ (`PI_CODING_AGENT_DIR`, `PI_CONFIG_DIR`, `XDG_DATA_HOME`, or a symlink). So a
50
+ writing sync also holds `.docks-kit-sync-lock.db` inside each such folder
51
+ that it writes. For `agents` this is the agents folder. For `omp` it is the
52
+ omp agent folder, the intercom folder, and the omp data root. The held line
60
53
  then names that lock file.
61
54
  - A writing sync creates a missing folder of this set just before it takes the
62
55
  lock of that folder, after the locks of its ancestors in the set. Each folder
@@ -81,29 +74,11 @@ same files at once.
81
74
  wait and the held line are those of a sync.
82
75
  - A run with `--dry-run`, `-h`, or `--help` takes no lock. A dry run through
83
76
  the `DRY_RUN=1` env global of the private `native-raw` engine channel takes
84
- no lock either. `model <tool>` without a value only reads, so it takes no
85
- lock.
77
+ no lock either.
86
78
  - The lock goes when the run ends, also when the process is killed.
87
79
  - A held lock also delays the error of an invalid argument until the wait
88
80
  ends.
89
81
 
90
- ## Per-tool flags
91
-
92
- | Flag | Effect |
93
- |------|--------|
94
- | `--claude-model=<m>` | Deploy-time modifier: deployed model (aliases or full claude-* IDs; `default` unsets) |
95
- | `--claude-effort=<level>` | Deploy-time modifier: `effortLevel`; valid `low`, `medium`, `high`, `xhigh`, or `default` (Claude SoT: `high`) |
96
- | `--claude-advisor=<state>` | Deploy-time modifier: advisor `on`, `off`, or `default` (SoT off/unset) |
97
- | `--claude-compact-window=<n>` | Deploy-time modifier: autocompact window in tokens (`680000` or `680k`) |
98
- | `--claude-permissive` | Deploy-time modifier: empty permissions.ask/deny (sandboxes) |
99
- | `--claude-plugin=<name>` | Sticky opt-in plugin (known: supabase, n8n); comma-separate for several |
100
- | `--codex-model=<m>` | Deploy-time modifier: deployed Codex model |
101
- | `--codex-effort=<level>` | Deploy-time modifier: `model_reasoning_effort`; valid `none`, `minimal`, `low`, `medium`, `high`, `xhigh`, `max`, `ultra`, or `default` (Codex SoT: `high`; model-dependent) |
102
-
103
- Bare model, effort, or advisor modifiers print the relevant valid-value catalog
104
- and exit 2. A modifier for a target not selected by the positional arguments is
105
- ignored with a warning; Claude modifiers never touch Codex config and vice versa.
106
-
107
82
  ## `omp` session flags
108
83
 
109
84
  | Flag | Effect |
@@ -164,10 +139,5 @@ Old flags exit with a rename hint — there is no compat behavior.
164
139
 
165
140
  | Old | New |
166
141
  |-----|-----|
167
- | `--claude` / `--codex` / `--agents` | positional `claude` / `codex` / `agents` |
168
- | `--force` | `--reconcile` |
142
+ | `--agents` / `--omp` | positional `agents` / `omp` |
169
143
  | `--remove-plugins` | `--prune` |
170
- | `--680k` | `--claude-compact-window=680k` |
171
- | `--permissive` | `--claude-permissive` |
172
- | `--supabase` / `--n8n` | `--claude-plugin=supabase` / `--claude-plugin=n8n` |
173
- | `--skip-rtk` | `--skip-bubblewrap` |
@@ -23,7 +23,7 @@ bypasses an ignored stale build.
23
23
  bun add -g docks-kit@latest
24
24
  docks-kit sync
25
25
  # or one-shot, no install:
26
- bunx docks-kit status
26
+ bunx docks-kit toolchain check
27
27
  ```
28
28
 
29
29
  The npm package bundles the CLI and its generated in-memory payload — releases
@@ -159,8 +159,7 @@ sync/config reads.
159
159
  ## Prerequisites
160
160
 
161
161
  - Bun for source/global installs; release binaries embed the runtime
162
- - Node/npm for npm-global LSP servers
163
- - rustup is optional and needed only for the Rust language server
162
+ - Node/npm for `npx` (the skills CLI) and for an npm-installed kit
164
163
  - jq is optional doctor/test tooling; sync has no jq runtime dependency
165
164
  - curl is needed only to download Bun's upstream installer when Bun is absent:
166
165
  by the POSIX `docks-kit` launcher, by the POSIX `install.sh`, and by sync's
@@ -33,8 +33,8 @@ constant.
33
33
 
34
34
  | Lane | Window | Reserve | Threshold | Percent of window |
35
35
  |---|---:|---:|---:|---:|
36
- | `openai-codex/gpt-6-astra`, `openai-codex/gpt-6-luna` | 272,000 | 40,800 | 231,200 | 85.0% |
37
- | `openai-codex/gpt-6.1-sol`, `openai-codex/gpt-6.1-terra` | 500,000 | 75,000 | 425,000 | 85.0% |
36
+ | `openai-codex/gpt-6.2-astra`, `openai-codex/gpt-6-luna` | 272,000 | 40,800 | 231,200 | 85.0% |
37
+ | `openai-codex/gpt-6.2-sol`, `openai-codex/gpt-6.1-terra` | 500,000 | 75,000 | 425,000 | 85.0% |
38
38
  | `anthropic/*` | 1,000,000 | 150,000 | 850,000 | 85.0% |
39
39
 
40
40
  Every lane compacts at the same fraction of its own window. The absolute
@@ -93,8 +93,8 @@ to `contextWindow - 1`. On any model whose window is at or below the pinned
93
93
  value, the reserve collapses to one token, so omp compacts only once the
94
94
  context is already full. omp 18.8.8 and later fall back to the reserve-based
95
95
  budget for such a pin, so this collapse does not occur there. The kit floor is
96
- omp 18.4.4, so a supported host can still have the clamp. The reserve-based
97
- default always leaves at least 15% of the window free.
96
+ omp 18.22.2, later than 18.8.8, so a supported host no longer has the clamp.
97
+ The reserve-based default always leaves at least 15% of the window free.
98
98
 
99
99
  No current role model has a window below 272,000, so the hazard was latent
100
100
  rather than live. Not pinning removes it entirely, including for any future
@@ -112,7 +112,7 @@ key turns idle compaction off on fresh installs.
112
112
 
113
113
  | Setting | Value | Reason |
114
114
  |---|---|---|
115
- | `extendedContext` | `false` | omp 17.4.0 added the setting with the default on, and a later release changed the default to off. The setting controls the window omp reports for a model with a long-context limit. On omp 18.10.4, `omp models openai-codex` lists `openai-codex/gpt-6.1-sol` at 500,000 tokens with the setting off and 922,000 tokens with it on. The Codex lane uses this model for the `default`, `task`, and `vision` fallbacks. The `reviewer` role uses `openai-codex/gpt-6.1-terra`, which omp 18.16.0 also lists at 500,000 tokens with the setting off and which has the same 272,000-token long-prompt rule. OpenAI bills the whole request at the long-prompt rate when the input is above 272,000 tokens. For GPT-6.1 Sol, that rate is input $4.00 instead of $2.00, cache read $0.20 instead of $0.10, cache write $5.00 instead of $2.50, and output $15.00 instead of $10.00. The cost is a cliff rather than a marginal rate. Keeping the setting off keeps the smaller window, so fewer requests cross that cliff. |
115
+ | `extendedContext` | `false` | omp 17.4.0 added the setting with the default on, and a later release changed the default to off. The setting controls the window omp reports for a model with a long-context limit. On omp 18.10.4, `omp models openai-codex` listed `openai-codex/gpt-6.1-sol` at 500,000 tokens with the setting off and 922,000 tokens with it on. The fallback chains now use `openai-codex/gpt-6.2-sol` for the `default`, `task`, and `vision` fallbacks. omp 18.22.2 lists that model at 500,000 tokens with the setting off. The kit has not checked the window that omp 18.22.2 reports for it with the setting on. The `reviewer` role uses `openai-codex/gpt-6.1-terra`, which omp 18.16.0 also lists at 500,000 tokens with the setting off and which has the same 272,000-token long-prompt rule. OpenAI bills the whole request at the long-prompt rate when the input is above 272,000 tokens. For GPT-6.2 Sol, OpenAI applies 2x the input and cache rates and 1.5x the output rate to the full request. From the base prices, that gives input $4.00 instead of $2.00, cache read $0.20 instead of $0.10, cache write $5.00 instead of $2.50, and output $15.00 instead of $10.00. The cost is a cliff rather than a marginal rate. Keeping the setting off keeps the smaller window, so fewer requests cross that cliff. |
116
116
  | `compaction.idleThresholdTokens` | `200000` | omp default. |
117
117
  | `compaction.idleTimeoutSeconds` | `300` | omp default. |
118
118
  | `compaction.methodOrder` | `["remote","snapcompact","handoff","shake","soft"]` | omp default, remote first. Since omp 18.1.18 the Anthropic server-side lane handles Anthropic models, so snapcompact is a fallback rather than the primary path. |
@@ -18,7 +18,7 @@ against.
18
18
  | `tiny` | `openai-codex/gpt-6-luna` | low | 21 | $0.0045 | 2.18 s |
19
19
  | `fable` | `anthropic/claude-fable-5-2` | medium | 56 (v5.0.1) | $3.80 (v5.0.1) | 11.23 s (v5.0.1) |
20
20
  | `switch_fable` | `anthropic/claude-fable-5-2` | medium | 56 (v5.0.1) | $3.80 (v5.0.1) | 11.23 s (v5.0.1) |
21
- | `astra` | `openai-codex/gpt-6-astra` | xhigh | 52 | $2.31 | 126.90 s |
21
+ | `astra` | `openai-codex/gpt-6.2-astra` | xhigh | 62 (v5.0.1) | $2.52 (v5.0.1) | 52.89 s (v5.0.1) |
22
22
  | `reviewer` | `openai-codex/gpt-6.1-terra` | high | 42 (v5.0.1) | $0.26 (v5.0.1) | 10.91 s (v5.0.1) |
23
23
  | `web` | `web/firecrawl` | n/a | n/a | n/a | n/a |
24
24
 
@@ -26,19 +26,38 @@ The table reports the measured Artificial Analysis figures for each assigned
26
26
  model and level. It states no motive that the config or omp's own
27
27
  documentation does not establish. AA has not measured output speed or latency
28
28
  for GPT-6 Luna medium, so that row carries `n/a` for TTFT. AA measures no web
29
- search provider, so the `web` row carries no figures. The `reviewer`, `fable`,
30
- and `switch_fable` rows come from later captures at Intelligence Index v5.0.1.
31
- The 2026-09-29 capture predates GPT-6.1 Terra and Fable 5.2. Compare those
32
- rows only with other v5.0.1 figures (see GPT-6.1 Terra and Claude Fable 5.2
33
- below). A review benchmark chose the model of the `reviewer` role
34
- (DocksDocks/public#130), not these figures. The tier of that role follows the
35
- main session (DocksDocks/public#134).
29
+ search provider, so the `web` row carries no figures. The `astra`, `reviewer`,
30
+ `fable`, and `switch_fable` rows come from later captures at Intelligence
31
+ Index v5.0.1: GPT-6.1 Terra from 2026-11-05, Fable 5.2 from 2026-11-12, and
32
+ GPT-6.2 Astra from 2026-11-18. The 2026-09-29 capture predates these models.
33
+ Compare those rows only with other v5.0.1 figures (see GPT-6.2 Astra, GPT-6.1
34
+ Terra, and Claude Fable 5.2 below). A review benchmark chose the model of the
35
+ `reviewer` role (DocksDocks/public#130), not these figures. The tier of that
36
+ role follows the main session (DocksDocks/public#134).
36
37
 
37
38
  The role map carries no Coding Agent Index column. That index measures specific
38
39
  agent harnesses and model levels, not omp roles. The snapshot section below
39
- lists the relevant entries, including Codex with GPT-6.1 Sol high.
40
+ lists the relevant entries, including Codex with GPT-6.2 Sol and GPT-6.2 Astra
41
+ at high and xhigh.
40
42
 
41
- ### GPT-6 Sol and GPT-6 Luna availability
43
+ ### AA rows for each role model
44
+
45
+ For each model that an omp role or retry chain uses, this topic records AA
46
+ rows at `high` and `xhigh` at least. Each set of rows names its Intelligence
47
+ Index version and its capture date. A role or chain change adds the missing
48
+ rows in the same commit. On 2026-11-18, the role map and the retry chains use
49
+ six models, and each has both rows:
50
+
51
+ | Model | Index version | Capture date |
52
+ |---|---|---|
53
+ | Claude Opus 5.5 | v4.3.2 | 2026-09-29 |
54
+ | GPT-6 Luna | v4.3.2 | 2026-09-29 |
55
+ | Claude Fable 5.2 | v5.0.1 | 2026-11-12 |
56
+ | GPT-6.1 Terra | v5.0.1 | 2026-11-05, speed re-measured 2026-11-18 |
57
+ | GPT-6.2 Sol | v5.0.1 | 2026-11-18 |
58
+ | GPT-6.2 Astra | v5.0.1 | 2026-11-18 |
59
+
60
+ ### OpenAI model availability in omp
42
61
 
43
62
  The `openai-codex` selectors for GPT-6 Sol and GPT-6 Luna were deployed on
44
63
  2026-09-22, while the rollout of both models was still in progress. The owner
@@ -74,6 +93,24 @@ and the request returned HTTP 401 `refresh_token_invalidated`. The Codex
74
93
  model cache was last fetched on 2026-09-22 and does not list `gpt-6.1-sol`,
75
94
  so that Codex run did not establish model availability.
76
95
 
96
+ On 2026-11-18, omp 18.22.2 listed `gpt-6.2-sol` with a 500K context window
97
+ and `gpt-6.2-astra` with a 272K context window. Both had low, medium, high,
98
+ xhigh, and max. An `omp -p --mode json` run on each selector at `:low`
99
+ recorded that same id as the serving model. omp 18.22.2 listed no GPT-6.2
100
+ Terra and no GPT-6.2 Luna. To recheck, run `omp models openai-codex`; the
101
+ `gpt-6.2-sol` and `gpt-6.2-astra` rows must appear. Sync deploys the selectors
102
+ without a catalog check. On an omp release without these ids, omp can resolve
103
+ them by fuzzy match to another model without a warning, as on 2026-09-22.
104
+
105
+ On 2026-11-19, omp 19.0.0 became the toolchain floor and verified pin. Its
106
+ `omp models --json` catalog listed each of the 12 model selectors of
107
+ `modelRoles` and `retry.fallbackChains` (the `web/` entries left out) as an
108
+ exact `provider/id` row, so no role depends on a fuzzy match. This check read
109
+ the catalog only; it ran no model request. The rows also carry the fields that
110
+ the free launcher reads (`ompOverlay.ts parseFreeModels`): a string
111
+ `selector`, a `cost` object with numeric `input` and `output`, a numeric
112
+ `contextWindow`, and a `thinking` ladder array or `null`.
113
+
77
114
  What omp's settings catalog establishes about these roles:
78
115
 
79
116
  - `cycleOrder` lists the roles the model switcher cycles. `fable` is the
@@ -94,7 +131,7 @@ provider. "OpenAI fast mode" below describes how that tier changes. The bundled
94
131
  `reviewer` and `security-reviewer` agents ship with omp, and the
95
132
  `plan-lifecycle` plugin ships `code-reviewer` and `plan-reviewer`.
96
133
  `retry.fallbackChains.reviewer` holds `anthropic/claude-opus-5-5:high`, and
97
- `retry.fallbackChains.task` holds `openai-codex/gpt-6.1-sol:high`; both are
134
+ `retry.fallbackChains.task` holds `openai-codex/gpt-6.2-sol:high`; both are
98
135
  cross-vendor fallbacks.
99
136
 
100
137
  The review benchmark of DocksDocks/public#130 ran on 2026-11-05. It used 24
@@ -117,10 +154,10 @@ recall difference down to -0.07 (bound -0.05). Both intervals include 0, and
117
154
  the user judged the difference as noise.
118
155
 
119
156
  `retry.fallbackChains.astra` holds `anthropic/claude-fable-5-2:medium`.
120
- `retry.fallbackChains.fable` holds `openai-codex/gpt-6-astra:xhigh`.
157
+ `retry.fallbackChains.fable` holds `openai-codex/gpt-6.2-astra:xhigh`.
121
158
  Each deliberate cycle stop falls to the other vendor. Without these explicit
122
159
  chains, `retry.fallbackChains.default` would send either stop to
123
- `openai-codex/gpt-6.1-sol:high`.
160
+ `openai-codex/gpt-6.2-sol:high`.
124
161
  Chain entries are concrete selectors, not role aliases, so this pair cannot
125
162
  recurse. The hidden `switch_fable` chain stays empty.
126
163
 
@@ -175,12 +212,13 @@ comes from the GPT-5.6 Sol high column of the GPT-6.1 Sol high comparison.
175
212
  Index composition changes between versions; figures from an earlier capture
176
213
  cannot be mixed with these.
177
214
 
178
- The Coding Agent Index page (`/agents/coding-agents`) lists 30 harness, model,
179
- and level entries. It covers GPT-6.1 Sol with Codex at low, medium, high,
180
- xhigh, and max; Opus 5.5 and Fable 5.1 with Claude Code at max; and Astra and
181
- Luna with Codex at max. It also covers Grok Build with Grok 4.7 at xhigh,
182
- Antigravity SDK with Gemini 3.8 Flash at high, and Opencode with GLM-5.3 and
183
- Kimi Code CLI with Kimi K3 without levels. The relevant published scores are:
215
+ At v1.5 on 2026-09-29, the Coding Agent Index page (`/agents/coding-agents`)
216
+ listed 30 harness, model, and level entries. It covered GPT-6.1 Sol with Codex
217
+ at low, medium, high, xhigh, and max; Opus 5.5 and Fable 5.1 with Claude Code
218
+ at max; and Astra and Luna with Codex at max. It also covered Grok Build with
219
+ Grok 4.7 at xhigh, Antigravity SDK with Gemini 3.8 Flash at high, and Opencode
220
+ with GLM-5.3 and Kimi Code CLI with Kimi K3 without levels. The relevant
221
+ published v1.5 scores are:
184
222
 
185
223
  | Harness and model | Coding Agent Index |
186
224
  |---|---:|
@@ -196,7 +234,8 @@ Kimi Code CLI with Kimi K3 without levels. The relevant published scores are:
196
234
  | Codex - GPT-6 Luna (max) | 41 |
197
235
 
198
236
  These are coding-agent scores for specific harnesses and levels, not scores
199
- for the omp role map.
237
+ for the omp role map. The v1.5 table stays as history. Coding Agent Index v2.1
238
+ below has the 2026-11-18 entries for GPT-6.2 Sol and GPT-6.2 Astra.
200
239
 
201
240
  Column meanings:
202
241
 
@@ -212,6 +251,29 @@ Column meanings:
212
251
  - **TB 4.0** - Terminal-Bench 4.0, one of the index evaluations (ten at v4.3.2,
213
252
  nine at v5.0.1).
214
253
 
254
+ ### Coding Agent Index v2.1
255
+
256
+ Source: `https://artificialanalysis.ai/agents/coding-agents`, read on
257
+ 2026-11-18 at Coding Agent Index v2.1. This version gives equal weight to
258
+ FrontierCode-AA 2.0, Terminal-Bench 4.1, and SWE-Atlas-QnA-AA, and lists 39
259
+ entries. The page draws the scores only in charts. The values below are the
260
+ `indexScore` fields of the page data, on a 0 to 1 scale. The cost and agent
261
+ wall time columns are the mean per task from the same data. This capture did
262
+ not observe the rendered integer values. Do not compare these values with the
263
+ v1.5 integers above.
264
+
265
+ | Harness and model | indexScore | Cost/task | Agent wall time s |
266
+ |---|---:|---:|---:|
267
+ | Codex - GPT-6.2 Astra (max) | 0.5347 | $5.117 | 1506.7 |
268
+ | Codex - GPT-6.2 Astra (xhigh) | 0.5312 | $3.198 | 992.4 |
269
+ | Codex - GPT-6.2 Astra (high) | 0.5228 | $2.372 | 947.4 |
270
+ | Codex - GPT-6.2 Sol (max) | 0.4862 | $1.216 | 1051.4 |
271
+ | Codex - GPT-6.2 Sol (xhigh) | 0.4979 | $0.898 | 755.8 |
272
+ | Codex - GPT-6.2 Sol (high) | 0.4746 | $0.764 | 635.4 |
273
+
274
+ GPT-6.2 Sol scores higher at `xhigh` than at `max`. v2.1 lists GPT-6.1 Sol
275
+ and GPT-6.1 Terra with Codex only at max.
276
+
215
277
  ### Claude Opus 5.5 (Anthropic) - `anthropic/claude-opus-5-5`
216
278
 
217
279
  | Level | Index | Cost/task | Tokens/task | Index tokens | Speed t/s | TTFT s | TB 4.0 |
@@ -294,7 +356,74 @@ baseline at v4.3.2.
294
356
  Price: $10.00 in, $50.00 out, $0.25 cache hit per 1M. Context 1M. All levels
295
357
  run with fallback.
296
358
 
297
- ### GPT-6 Astra (OpenAI) - `openai-codex/gpt-6-astra`
359
+ ### GPT-6.2 Sol (OpenAI) - `openai-codex/gpt-6.2-sol`
360
+
361
+ Source: `https://artificialanalysis.ai`, read on 2026-11-18 at Intelligence
362
+ Index v5.0.1. The rows come from the rendered metric table of
363
+ `/models/comparisons/gpt-6-2-sol-vs-gpt-6-1-terra`, which shows every level of
364
+ both models. Tokens per task and index tokens come from the data of the model
365
+ page `/models/gpt-6-2-sol`, because the comparison table does not show them.
366
+ The other values in that data match the rendered table. These rows compare
367
+ only with other v5.0.1 figures, not with the v4.3.2 rows.
368
+
369
+ | Level | Index | Cost/task | Tokens/task | Index tokens | Speed t/s | TTFT s | TB 4.0 |
370
+ |---|---:|---:|---:|---:|---:|---:|---:|
371
+ | max | 57 | $1.01 | 40k | 61M | 85 | 66.70 | 60% |
372
+ | xhigh | 56 | $0.65 | 24k | 33M | 81 | 26.63 | 56% |
373
+ | high | 55 | $0.56 | 20k | 23M | 78 | 9.63 | 59% |
374
+
375
+ AA and OpenAI list $2.00 input, $0.10 cached input, $2.50 cache write, and
376
+ $10.00 output per 1M tokens. OpenAI also lists a 1,050,000-token context,
377
+ 922,000 maximum input tokens, and 128,000 maximum output tokens
378
+ (<https://developers.openai.com/api/docs/models/gpt-6.2-sol>). Prompts above
379
+ 272K input tokens cost 2x input and cache rates and 1.5x output rates for the
380
+ full request. The knowledge cutoff is 2026-06-01. OpenAI released the model on
381
+ 2026-11-17 (<https://developers.openai.com/api/docs/changelog>). Codex uses a
382
+ 500,000-token default context (<https://developers.openai.com/codex/models>).
383
+ On 2026-11-18, omp 18.22.2 listed `gpt-6.2-sol` with the same 500K context
384
+ window, 128K maximum output, and the levels low, medium, high, xhigh, and max.
385
+
386
+ `retry.fallbackChains.default` and `.task` moved from GPT-6.1 Sol `high` to
387
+ GPT-6.2 Sol `high`, and `.vision` moved at `medium` (DocksDocks/public#175).
388
+ On 2026-11-18, the v5.0.1 data of the model page `/models/gpt-6-1-sol` listed
389
+ GPT-6.1 Sol high at index 50 and $0.59 per task. GPT-6.2 Sol high has 55 and
390
+ $0.56.
391
+
392
+ ### GPT-6.2 Astra (OpenAI) - `openai-codex/gpt-6.2-astra`
393
+
394
+ Source: `https://artificialanalysis.ai`, read on 2026-11-18 at Intelligence
395
+ Index v5.0.1. The rows come from the rendered metric table of
396
+ `/models/comparisons/gpt-6-2-astra-vs-gpt-6-2-sol`. Tokens per task and index
397
+ tokens come from the data of the model page `/models/gpt-6-2-astra`. These
398
+ rows compare only with other v5.0.1 figures, not with the v4.3.2 rows.
399
+
400
+ | Level | Index | Cost/task | Tokens/task | Index tokens | Speed t/s | TTFT s | TB 4.0 |
401
+ |---|---:|---:|---:|---:|---:|---:|---:|
402
+ | max | 62 | $4.00 | 41k | 66M | 53 | 170.99 | 64% |
403
+ | xhigh | 62 | $2.52 | 22k | 27M | 54 | 52.89 | 64% |
404
+ | high | 60 | $2.07 | 18k | 20M | 56 | 22.07 | 60% |
405
+
406
+ The `max` and `xhigh` rows both round to 62. The page data gives 62.14 and
407
+ 61.56.
408
+
409
+ AA and OpenAI list $8.00 input, $0.40 cached input, $10.00 cache write, and
410
+ $40.00 output per 1M tokens. OpenAI also lists a 1,050,000-token context,
411
+ 922,000 maximum input tokens, and 128,000 maximum output tokens
412
+ (<https://developers.openai.com/api/docs/models/gpt-6.2-astra>). The
413
+ long-prompt rule of GPT-6.2 Sol also applies. The knowledge cutoff is
414
+ 2026-05-18. OpenAI released the model on 2026-11-17, in the same changelog
415
+ entry as GPT-6.2 Sol. Codex uses a 272,000-token default context. On
416
+ 2026-11-18, omp 18.22.2 listed `gpt-6.2-astra` with the same 272K context
417
+ window, 128K maximum output, and the levels low, medium, high, xhigh, and max.
418
+
419
+ `modelRoles.astra` and `retry.fallbackChains.fable` moved from GPT-6 Astra
420
+ `xhigh` to GPT-6.2 Astra `xhigh` (DocksDocks/public#175). GPT-6 Astra below
421
+ has the v5.0.1 figures for a like-for-like comparison.
422
+
423
+ ### GPT-6 Astra (OpenAI) - previous generation
424
+
425
+ The role map and the retry chains no longer use GPT-6 Astra. This table stays
426
+ as the v4.3.2 baseline for the previous `astra` role.
298
427
 
299
428
  | Level | Index | Cost/task | Tokens/task | Index tokens | Speed t/s | TTFT s | TB 4.0 |
300
429
  |---|---:|---:|---:|---:|---:|---:|---:|
@@ -307,7 +436,17 @@ run with fallback.
307
436
  Price: $10.00 in, $50.00 out, $1.00 cache hit per 1M. Context 1M. Knowledge
308
437
  cutoff 2026-04-30. AA publishes no non-reasoning Astra row.
309
438
 
310
- ### GPT-6.1 Sol (OpenAI) - `openai-codex/gpt-6.1-sol`
439
+ On 2026-11-18, the v5.0.1 data of the AA model page `/models/gpt-6-astra`
440
+ listed GPT-6 Astra `xhigh` at index 55 and $3.66 per task, with 44 t/s and an
441
+ 83.77 s TTFT. It listed `high` at index 55, $3.24, 44 t/s, and 40.21 s. At
442
+ `xhigh`, GPT-6.2 Astra scores 62 against 55 and costs $2.52 against $3.66 per
443
+ task. It produces 54 against 44 t/s, with a TTFT of 52.89 s against 83.77 s.
444
+ AA marks GPT-6 Astra deprecated, so its comparison pages return HTTP 404.
445
+
446
+ ### GPT-6.1 Sol (OpenAI) - previous generation
447
+
448
+ The role map and the retry chains no longer use GPT-6.1 Sol. This table stays
449
+ as the v4.3.2 baseline for the former `task` role and retry fallbacks.
311
450
 
312
451
  | Level | Index | Cost/task | Tokens/task | Index tokens | Speed t/s | TTFT s | TB 4.0 |
313
452
  |---|---:|---:|---:|---:|---:|---:|---:|
@@ -323,7 +462,9 @@ OpenAI also lists $2.50 per 1M cache-write tokens, a 1,050,000-token context,
323
462
  272K input tokens cost 2x input and cache rates and 1.5x output rates for the
324
463
  full request. The knowledge cutoff is 2026-04-30. The API effort ladder is
325
464
  `low, medium, high, xhigh, max`, with `medium` as the default. There is no
326
- `none`, `minimal`, or non-reasoning level.
465
+ `none`, `minimal`, or non-reasoning level. On 2026-11-18, omp 18.22.2 still
466
+ served `gpt-6.1-sol`, and the OpenAI deprecations page did not list it. AA
467
+ marks it deprecated.
327
468
 
328
469
  ### GPT-6.1 Terra (OpenAI) - `openai-codex/gpt-6.1-terra`
329
470
 
@@ -346,6 +487,14 @@ For comparison at the same index version, AA lists GPT-6.1 Sol high at index
346
487
  TTFT, and 52% on Terminal-Bench 4.0. The same Sol high row read 50, $0.32, and
347
488
  57.26 s at v4.3.2, which shows why the two captures do not mix.
348
489
 
490
+ On 2026-11-18, the release comparison page
491
+ `/models/comparisons/gpt-6-2-sol-vs-gpt-6-1-terra` showed new speed and TTFT
492
+ figures for GPT-6.1 Terra at v5.0.1. `high` ran at 65 t/s with a 10.50 s
493
+ TTFT, `xhigh` at 65 t/s with 30.33 s, and `max` at 69 t/s with 146.49 s. The
494
+ index, cost, and token figures were unchanged. AA measures speed and latency
495
+ continuously, so the table keeps the 2026-11-05 rows. The speed comparison
496
+ below uses the 2026-11-18 figures.
497
+
349
498
  AA and OpenAI list $1.00 input, $5.00 output, and $0.05 cached input per 1M
350
499
  tokens. OpenAI also lists a 1,050,000-token context, 922,000 maximum input
351
500
  tokens, and 128,000 maximum output tokens
@@ -353,6 +502,30 @@ tokens, and 128,000 maximum output tokens
353
502
  Fast mode costs 2x Standard. The knowledge cutoff is 2026-06-01. The API effort
354
503
  ladder is `none, low, medium, high, xhigh, max`, with `medium` as the default.
355
504
 
505
+ ### Output speed: GPT-6.2 Sol and GPT-6.1 Terra
506
+
507
+ GPT-6.2 Sol has the higher output speed at `high` and at `xhigh`.
508
+ The figures come from one page,
509
+ `/models/comparisons/gpt-6-2-sol-vs-gpt-6-1-terra`, read on 2026-11-18 at
510
+ Intelligence Index v5.0.1.
511
+
512
+ | Metric | GPT-6.2 Sol | GPT-6.1 Terra |
513
+ |---|---:|---:|
514
+ | `high` output speed | 78 t/s | 65 t/s |
515
+ | `xhigh` output speed | 81 t/s | 65 t/s |
516
+ | `high` TTFT | 9.63 s | 10.50 s |
517
+ | `xhigh` TTFT | 26.63 s | 30.33 s |
518
+ | `high` end-to-end time | 42.65 s | 43.91 s |
519
+ | `xhigh` end-to-end time | 85.03 s | 86.53 s |
520
+
521
+ At `high`, Sol produces about 19% more tokens per second than Terra. At
522
+ `xhigh`, the gain is about 24%. Sol also has the lower TTFT and end-to-end
523
+ time at both levels.
524
+
525
+ The `reviewer` role stays on GPT-6.1 Terra `high`. The #130 benchmark chose
526
+ Terra by escaped defects, not by speed. omp 18.22.2 lists no GPT-6.2 Terra. A
527
+ move of the reviewer to GPT-6.2 Sol needs a new benchmark.
528
+
356
529
  ### GPT-6 Luna (OpenAI) - `openai-codex/gpt-6-luna`
357
530
 
358
531
  | Level | Index | Cost/task | Tokens/task | Index tokens | Speed t/s | TTFT s | TB 4.0 |
@@ -373,7 +546,7 @@ has measured both at the other levels.
373
546
  ### GPT-5.6 Sol (OpenAI) - previous generation
374
547
 
375
548
  The role map no longer uses GPT-5.6 Sol. This table stays as the measured
376
- baseline for the GPT-6.1 Sol `task` retry fallback.
549
+ baseline for the former GPT-6.1 Sol `task` retry fallback.
377
550
 
378
551
  | Level | Index | Cost/task | Tokens/task | Index tokens | Speed t/s | TTFT s | TB 4.0 |
379
552
  |---|---:|---:|---:|---:|---:|---:|---:|
@@ -407,12 +580,16 @@ Price: $0.20 in, $1.20 out, $0.02 cache hit per 1M. Context 1M.
407
580
 
408
581
  `task` runs `anthropic/claude-opus-5-5:high`, the same model and level as
409
582
  `default`. The owner moved it from `openai-codex/gpt-6.1-sol:high` on
410
- 2026-10-09. GPT-6.1 Sol high is now the `task` retry fallback. The owner uses
411
- Astra only for main orchestration, so Astra has a dedicated `astra` cycle stop
412
- at `xhigh`. This comparison keeps the retired Astra-low, GPT-5.6 Sol, and
413
- GPT-6.1 Sol choices beside the current Opus level.
414
-
415
- | Metric | Astra low, retired | GPT-5.6 Sol high, retired | GPT-6.1 Sol high, fallback | GPT-6.1 Sol max | Opus 5.5 high, current |
583
+ 2026-10-09. GPT-6.1 Sol high then became the `task` retry fallback. Since
584
+ DocksDocks/public#175, the fallback is GPT-6.2 Sol high
585
+ (`openai-codex/gpt-6.2-sol:high`). The owner uses Astra only for main
586
+ orchestration, so Astra has a dedicated `astra` cycle stop at `xhigh`. This
587
+ comparison keeps the retired Astra-low, GPT-5.6 Sol, and GPT-6.1 Sol choices
588
+ beside the current Opus level. It is the v4.3.2 capture of 2026-09-29, so it
589
+ does not compare with the v5.0.1 GPT-6.2 rows. GPT-6.2 Sol above has the
590
+ figures for the current fallback.
591
+
592
+ | Metric | Astra low, retired | GPT-5.6 Sol high, retired | GPT-6.1 Sol high, former fallback | GPT-6.1 Sol max | Opus 5.5 high, current |
416
593
  |---|---:|---:|---:|---:|---:|
417
594
  | Intelligence Index | 46 | 42 | 50 | 52 | 54 |
418
595
  | Cost per Index task | $0.82 | $0.81 | $0.32 | $0.72 | $1.82 |
@@ -443,9 +620,9 @@ Opus 5.5 high scores 54 against GPT-6.1 Sol high at 50, 57% against 52% on
443
620
  Terminal-Bench 4.0, and 1705 against 1471 on AA-Briefcase. It costs $1.82
444
621
  against $0.32 per index task. Its answer TTFT is 52.85 s against 57.26 s.
445
622
 
446
- The `astra` cycle stop runs xhigh: index 52, $2.31 per index task, and
447
- 126.90 s TTFT. AA publishes a Coding Agent Index entry for Astra only at
448
- max with Codex, where it scores 62.
623
+ The `astra` cycle stop runs GPT-6.2 Astra `xhigh`. At v5.0.1 it has index 62,
624
+ $2.52 per index task, and a 52.89 s TTFT (see GPT-6.2 Astra). Coding Agent
625
+ Index v2.1 lists Codex with GPT-6.2 Astra at high, xhigh, and max.
449
626
 
450
627
  ## How Astra is selected in practice
451
628
 
@@ -456,10 +633,13 @@ No subagent selects Astra automatically.
456
633
 
457
634
  `SoT/.omp/models.yml` declares the full `low, medium, high, xhigh, max` ladder
458
635
  with `defaultLevel: xhigh` under
459
- `providers.openai-codex.modelOverrides.gpt-6-astra.thinking`.
636
+ `providers.openai-codex.modelOverrides.gpt-6.2-astra.thinking`.
460
637
  The override stays as the kit's worked example of a provider ladder
461
638
  redefinition. The in-session thinking control can still select `low` for a
462
- quick answer.
639
+ quick answer. The block is keyed by model id, so it moved from `gpt-6-astra`
640
+ to `gpt-6.2-astra` with the role. `ompRemovals.ts syncOmpModelRemovals`
641
+ prunes the deployed `gpt-6-astra` block while it equals the shipped block, and
642
+ keeps a block that the user edited.
463
643
 
464
644
  - The bundled `scout` and `sonic` agents carry `model: "@smol"` and
465
645
  `thinking-level: medium` in their embedded frontmatter, so they run Luna,
@@ -493,9 +673,12 @@ The runtime per-agent measurement from fresh `omp -p` runs on 2026-09-09
493
673
  predates this change. It applies to the retired Astra-low `task` configuration,
494
674
  not the current role map.
495
675
 
496
- omp's 272k context window is the `/extended-context off` window for Astra.
497
- `/extended-context on` uses the 922k input window, 1.05M total, which matches
498
- AA's 1M.
676
+ On 2026-11-18, omp 18.22.2 listed GPT-6.2 Astra with a 272,000-token context
677
+ window. This is the `/extended-context off` window, and it equals the Codex
678
+ default context. OpenAI lists a 1,050,000-token context and 922,000 maximum
679
+ input tokens. For GPT-6 Astra, `/extended-context on` used the 922k input
680
+ window. The kit has not checked the window that `/extended-context on` gives
681
+ for GPT-6.2 Astra.
499
682
 
500
683
  ## Free session launcher (`docks-kit omp`)
501
684
 
@@ -680,10 +863,13 @@ and the ladder counts in these docs in the same commit.
680
863
  values, and exits without starting a session. `priority` is the tier omp's
681
864
  `/fast on` selects for OpenAI and OpenAI-Codex models. `--fast` over
682
865
  `ultrafast` selects `priority`. The kit ships no Ultrafast toggle: omp honors
683
- `ultrafast` for a Codex model only when its catalog `serviceTiers` lists it,
684
- and the omp 18.4.9 catalog lists none for GPT-6.1 Sol, Astra, or Luna. The SoT
685
- does not declare `tier.openai`, so the additive sync merge keeps the toggled
686
- value. New sessions read it; a running session changes through `/fast`.
866
+ `ultrafast` for a Codex model only when its catalog `serviceTiers` lists it.
867
+ The omp 18.4.9 catalog listed none for GPT-6.1 Sol, Astra, or Luna. On
868
+ 2026-11-18, `omp models openai-codex --json` on omp 18.22.2 listed no
869
+ `serviceTiers` for any `gpt-6` id, `gpt-6.2-sol` and `gpt-6.2-astra`
870
+ included. The SoT does not declare `tier.openai`, so the additive sync merge
871
+ keeps the toggled value. New sessions read it; a running session changes
872
+ through `/fast`.
687
873
 
688
874
  The four review agents inherit the live tier of the main session
689
875
  (`task.agentServiceTierOverrides: inherit`). In a new session after
@@ -702,12 +888,27 @@ on normal processing, use `docks-kit omp --fast` and a new session, not
702
888
 
703
889
  ## Maintenance
704
890
 
705
- - Refresh each per-level row from the metric table of
706
- `/models/comparisons/<level-slug>-vs-gpt-5-6-sol-high`. Check that the page
707
- title names the requested model and level. Take `max` from the bare model
708
- slug. Do not read values from the chart payloads of the model pages: each
709
- chart holds only about 20 models, so a missing value there does not mean
710
- that AA has not measured it.
891
+ - Refresh each per-level row from the rendered metric table of a release
892
+ comparison page, for example
893
+ `/models/comparisons/gpt-6-2-sol-vs-gpt-6-1-terra` and
894
+ `/models/comparisons/gpt-6-2-astra-vs-gpt-6-2-sol`. Each page shows every
895
+ level of both models. Take tokens per task and index tokens from the data
896
+ of the model page, for example `/models/gpt-6-2-sol`, and check that its
897
+ other values match the rendered table. A model-page chart holds only about
898
+ 20 models, so a value missing there does not mean that AA has not measured
899
+ it.
900
+ - On 2026-11-18, the per-level pattern
901
+ `/models/comparisons/<level-slug>-vs-gpt-5-6-sol-high` returned HTTP 404. A
902
+ comparison with a model that AA marks deprecated also returned HTTP 404, for
903
+ example `gpt-6-1-sol`, `gpt-6-astra`, or `gpt-5-6-sol`. Read such a model
904
+ from the data of its model page.
905
+ - Keep AA rows at `high` and `xhigh` at least for each model that an omp role
906
+ or retry chain uses, with the index version and the capture date (see AA
907
+ rows for each role model). Add the missing rows in the same commit as a
908
+ role or chain change.
909
+ - After a role or chain change, run `omp models openai-codex` and check that
910
+ it lists every `openai-codex` id that the roles and chains name. Sync does
911
+ not check the omp catalog.
711
912
  - Refresh the Coding Agent Index from `/agents/coding-agents`.
712
913
  - Record the index version with the numbers. AA changes index composition
713
914
  between versions, so a score from another version is not a comparison.