docks-kit 0.22.23 → 0.22.28

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.
package/AGENTS.md CHANGED
@@ -69,7 +69,7 @@ Module map for `cli/` (engine axes, commands, kit store, test fixtures, and the
69
69
 
70
70
  Codex SoT notes:
71
71
  - `SoT/.codex/AGENTS.md` deploys to `~/.codex/AGENTS.md` as global Codex instructions.
72
- - `SoT/.codex/config.toml` pins Codex to `model = "gpt-6.1-sol"`, sets normal and plan reasoning to `high` with concise summaries, and sets `model_verbosity = "low"`, live top-level `web_search`, workspace-write sandboxing with sandboxed command network access, cross-session `memories` (+ dedicated note tools), `[agents]` subagent limits (`max_threads = 12`, `max_depth = 2` — intentionally above Codex defaults for broad parallel kit work; deeper recursion increases cost and predictability risk), a 128 KiB `project_doc_max_bytes` budget for the repo-side AGENTS.md chain (the global `~/.codex/AGENTS.md` is uncapped and not counted), and enables the two Docks plugins `docks@docks` and `plan-lifecycle@docks` (the shared plan lifecycle).
72
+ - `SoT/.codex/config.toml` pins Codex to `model = "gpt-6.2-sol"`, sets normal and plan reasoning to `high` with concise summaries, and sets `model_verbosity = "low"`, live top-level `web_search`, workspace-write sandboxing with sandboxed command network access, cross-session `memories` (+ dedicated note tools), `[agents]` subagent limits (`max_threads = 12`, `max_depth = 2` — intentionally above Codex defaults for broad parallel kit work; deeper recursion increases cost and predictability risk), a 128 KiB `project_doc_max_bytes` budget for the repo-side AGENTS.md chain (the global `~/.codex/AGENTS.md` is uncapped and not counted), and enables the two Docks plugins `docks@docks` and `plan-lifecycle@docks` (the shared plan lifecycle). The two project Codex reviewer agents in `.codex/agents/` use the same model.
73
73
  - `SoT/.codex/config.toml` does not set `personality`. Codex 0.159.0 deprecated the key, and `friendly` and `pragmatic` no longer select a style. `codexConfig.ts removeRetiredTopLevelSettings, retired top-level key prune` removes the retired `personality = "pragmatic"` line from a deployed `config.toml` on every sync. A different user-set value survives.
74
74
  - `SoT/.codex/rules/*.rules` deploys to `~/.codex/rules/` as kit-managed Codex command policy. This is Codex's equivalent of permission allow/prompt/block rules; user-learned approvals in `~/.codex/rules/default.rules` are preserved. Codex applies the strictest matching rule across all files, so a user allow rule cannot loosen a kit `forbidden` or `prompt` rule. Because the strictest rule wins, a narrow allow cannot carve an exception out of a broader prompt prefix, and a prefix rule cannot see an option after an operand. The guards therefore fail closed: every `rm`, `git rm`, `git worktree remove`, and `gio trash` prompts, and only macOS `trash` is allowed (owner decision). A `prompt` goes to `approvals_reviewer = "auto_review"`, a reviewer subagent, not always to a person. `docks.rules` carries the destructive-command guards (`docks-kit docs command-guards`); `cli/test/unit/codexSyncRules.test.ts` parses every one-line `prefix_rule`, alternatives lists and `justification` included, and asserts each guard's decision. Keep each rule on one line so that parser sees it.
75
75
  - `codexSync.ts syncRules, kit rules manifest` records the sha256 of the kit bytes of each rules file that the kit deploys in `~/.codex/.kit-managed-rules`, in `<sha256> <file name>` lines. The manifest sits outside `rules/`, so Codex never reads it as a rule file. Sync writes each rules file through `atomicWrite.ts replaceFile`, so an interrupted write keeps the previous file. Codex loads only `*.rules` files, so it never loads the `*.tmp` sibling. Before the deploy, sync writes the manifest ahead. It keeps the recorded set of hashes of each name and adds the new kit hash of each shipped name. So a name whose kit bytes change has two lines. A sync that stops after a deploy thus leaves a record of the bytes that it wrote. A stale kit rules file keeps its old rules active. When the SoT drops a rules file, every sync therefore removes the deployed copy while its hash is in the recorded set of its name. A dropped file that the user edited stays, keeps its set, and gets a warning on each sync. Every recorded hash is a hash of kit bytes, so no record lets cleanup remove bytes that the user wrote. Cleanup reads and removes only recorded names that the SoT no longer ships. It never reads or removes a file that the SoT does not ship and no sync recorded, such as `default.rules`. Sync still deploys each name that the SoT ships over an existing file, recorded or not. It backs up a file that differs to `<name>.bak` and replaces it. After the deploy and the cleanup, the final manifest holds one line per shipped name, plus the set of each kept dropped name. Sync replaces the manifest through a temporary file and a rename, so an interrupted sync keeps the previous complete manifest. `--dry-run` prints `[dry-run] remove <file>` and writes nothing.
@@ -84,22 +84,22 @@ Codex SoT notes:
84
84
  - Claude permission rules are pruned, not host-scoped. The kit deploys its `PowerShell(...)` deny and ask rules on every host, because Claude Code leaves the PowerShell tool opt-in off Windows rather than unavailable, and a host that enables it must already carry the guards. `claudeRetired.ts RETIRED_PERMISSION_RULES, exact retired-rule inventory` lists the rules the kit no longer ships. `claudeRemovals.ts syncRemovals, retired-permission pass` force-prunes those exact strings from the kit-managed `~/.claude/settings.json` on every sync, because `settings.ts mergeSettings, permission-array union` would otherwise keep a dropped rule forever. A different user-authored rule survives. To restore an exact retired rule for one checkout, put it in that checkout's `.claude/settings.local.json`. Claude Code resolves that file against the working directory and merges it over user settings. Sync never reads or writes the checkout-local file. Claude Code has no user-scope local settings file. For a machine-wide restoration, add the rule to `SoT/.claude/settings.json`. At the same time, remove the exact string from `claudeRetired.ts RETIRED_PERMISSION_RULES, exact retired-rule inventory`. `permissions.allow` holds four read-only entries, `Read`, `Glob`, `Grep`, and `WebSearch`, the working-directory edit rule `Edit(./)`, and two narrow trash allows, `Bash(gio trash -- *)` and `Bash(trash *)`. Broad shell allow rules are not restored, because an allow rule resolves before Claude Code's read-only command analyzer and the auto-mode classifier. `autoMemoryEnabled` and `autoDreamEnabled` are both `false`.
85
85
 
86
86
  omp SoT notes:
87
- - `SoT/.omp/AGENTS.md`, `config.yml`, `models.yml`, and `mcp.json` deploy to `~/.omp/agent/`.
87
+ - `SoT/.omp/AGENTS.md`, `config.yml`, `models.yml`, and `mcp.json` deploy to `~/.omp/agent/`. omp loads `~/.omp/agent/AGENTS.md` only into main sessions and advisors, never into task subagents (`omp://context-files.md`). So `SoT/.omp/AGENTS.md` holds only main-session rules, and a command rule that subagents need goes into a kit plugin trigger rule.
88
88
  - `SoT/.omp/intercom.json` deploys to `$PI_CODING_AGENT_DIR/intercom/config.json`. The default root is `~/.pi/agent`. Under a named omp profile (`OMP_PROFILE` or `PI_PROFILE`), it deploys to `<profile agentDir>/intercom/config.json`, because omp exports `PI_CODING_AGENT_DIR` as the profile agent directory.
89
89
  - `ompFileDeploy.ts syncMergedYaml` deep-merges `config.yml` through `ompYaml.ts mergeOmpConfig` and `models.yml` through `mergeOmpModels`. Both wrap one generic mapping merge; only the config wrapper prunes stale `retry.fallbackChains` wildcards and merges `bash.patterns`.
90
- - `SoT/.omp/config.yml` `bash.patterns` holds the destructive-command guards. omp's `CRITICAL_BASH_PATTERNS` check returns a bare safety override before a matching prompt rule, and omp's default `tools.approvalMode: yolo` ignores a bare override, so a prompt on a critical command becomes an allow, in subagents too (`omp://approval-mode.md`). Every `rm` shape that check can flag is therefore a deny: recursive `rm`, `sudo` anywhere before `rm` (also by path), any `rm` with an `r` or `f` option before an absolute path, and `--no-preserve-root`. By owner decision (0.20.9), the recursive-`rm` deny also covers any prefix (a `* rm ` copy of each bare cluster rule, so `env X=1 rm -rf b`, `xargs rm -rf`, `find . -exec rm -rf {} +`, and `git rm -r dir` are denied) and the `-ri`, `-ir`, `-Ri`, and `-iR` clusters for bare `rm`, `*/rm`, and `* rm`. By owner decision (0.21.0), 17 more denies cover bulk removal that can empty the home directory: `find -delete` (bare, `*/find`, `* find`), `-exec`, `-execdir`, `-ok`, and `-okdir` with `rm` or `*/rm`, and `xargs rm` in six shapes. No glob can express "inside the temporary directory", so `docks-kit scratch remove` checks that path in code. A `$` or a backtick anywhere after `git push` in the same command is a deny, not a prompt (0.22.5, #120). A variable can expand to a `+` refspec or `--force`, and the matcher reads the text before the shell expands it. The agent reads the refusal and writes the ref literally, so no person is asked. The two rules are `*git push *$*` and ``*git push *`*``. Their leading `*` keeps a shell keyword such as `do`, a wrapper such as `env`, or a path before `git` from hiding the push. By owner decision (0.22.8, #126), the kit ships no omp prompt rule: `bash.patterns` holds only the denies plus the `trash *` allow, so every command that no deny matches runs without a question, in subagents too. Since 0.20.9 this run-without-question behavior covers a non-denied `rm`, `git rm`, `git worktree remove`, and `gio trash`. Since 0.21.0 it also covers `docks-kit scratch remove`. Since 0.22.8 it also covers `git clean`, `git reset --hard`, a hook skip, and every `git push` that no deny matches. The force-push denies match only `git push --force` or `-f` directly before `origin main` or `origin master`, so another spelling, such as `git push origin main --force`, runs. A prompt rule that the user adds survives sync and still asks. `SoT/.omp/AGENTS.md` carries the agent rules for these commands, and it tells the agent to write each `git push` ref literally and to push the current branch with `git push -u origin HEAD`. Claude Code `permissions.ask` and Codex `docks.rules` keep their ask tiers. The list order is deny, then `trash *`. omp takes the first match. `ompYaml.ts mergeBashPatterns, grouped rule merge` keeps user rules and groups all rules as kit deny, user deny, user prompt, kit prompt, other user rules, kit allow, so a user deny or prompt precedes every kit allow. It treats a deployed rule equal to a kit rule (approval case-folded, `match` whitespace collapsed) as the kit copy, so a re-sync is a no-op. A rule the kit stops shipping stays as a user rule, so retire one through `ompRemovals.ts OMP_RETIRED_BASH_RULES`. `cli/test/lib/ompBashPatterns.ts` ports omp 18.17.0's matcher, critical patterns, and yolo resolution; `cli/test/unit/ompBashPatterns.test.ts` asserts each command's final outcome and the merge. omp subagents reject every `prompt`, and the `eval` tool stays unrestricted by owner decision.
90
+ - `SoT/.omp/config.yml` `bash.patterns` holds the destructive-command guards. omp's `CRITICAL_BASH_PATTERNS` check returns a bare safety override before a matching prompt rule, and omp's default `tools.approvalMode: yolo` ignores a bare override, so a prompt on a critical command becomes an allow, in subagents too (`omp://approval-mode.md`). Every `rm` shape that check can flag is therefore a deny: recursive `rm`, `sudo` anywhere before `rm` (also by path), any `rm` with an `r` or `f` option before an absolute path, and `--no-preserve-root`. By owner decision (0.20.9), the recursive-`rm` deny also covers any prefix (a `* rm ` copy of each bare cluster rule, so `env X=1 rm -rf b`, `xargs rm -rf`, `find . -exec rm -rf {} +`, and `git rm -r dir` are denied) and the `-ri`, `-ir`, `-Ri`, and `-iR` clusters for bare `rm`, `*/rm`, and `* rm`. By owner decision (0.21.0), 17 more denies cover bulk removal that can empty the home directory: `find -delete` (bare, `*/find`, `* find`), `-exec`, `-execdir`, `-ok`, and `-okdir` with `rm` or `*/rm`, and `xargs rm` in six shapes. No glob can express "inside the temporary directory", so `docks-kit scratch remove` checks that path in code. A `$` or a backtick anywhere after `git push` in the same command is a deny, not a prompt (0.22.5, #120). A variable can expand to a `+` refspec or `--force`, and the matcher reads the text before the shell expands it. The agent reads the refusal and writes the ref literally, so no person is asked. The two rules are `*git push *$*` and ``*git push *`*``. Their leading `*` keeps a shell keyword such as `do`, a wrapper such as `env`, or a path before `git` from hiding the push. By owner decision (0.22.8, #126), the kit ships no omp prompt rule: `bash.patterns` holds only the denies plus the `trash *` allow, so every command that no deny matches runs without a question, in subagents too. Since 0.20.9 this run-without-question behavior covers a non-denied `rm`, `git rm`, `git worktree remove`, and `gio trash`. Since 0.21.0 it also covers `docks-kit scratch remove`. Since 0.22.8 it also covers `git clean`, `git reset --hard`, a hook skip, and every `git push` that no deny matches. The force-push denies match only `git push --force` or `-f` directly before `origin main` or `origin master`, so another spelling, such as `git push origin main --force`, runs. A prompt rule that the user adds survives sync and still asks. The kit plugin trigger rules in `SoT/.omp/plugin/rules/` carry the agent rules for these commands. `git-safety.md` tells the agent to write each `git push` ref literally and to push the current branch with `git push -u origin HEAD`. Claude Code `permissions.ask` and Codex `docks.rules` keep their ask tiers. The list order is deny, then `trash *`. omp takes the first match. `ompYaml.ts mergeBashPatterns, grouped rule merge` keeps user rules and groups all rules as kit deny, user deny, user prompt, kit prompt, other user rules, kit allow, so a user deny or prompt precedes every kit allow. It treats a deployed rule equal to a kit rule (approval case-folded, `match` whitespace collapsed) as the kit copy, so a re-sync is a no-op. A rule the kit stops shipping stays as a user rule, so retire one through `ompRemovals.ts OMP_RETIRED_BASH_RULES`. `cli/test/lib/ompBashPatterns.ts` ports omp 18.17.0's matcher, critical patterns, and yolo resolution; `cli/test/unit/ompBashPatterns.test.ts` asserts each command's final outcome and the merge. omp subagents reject every `prompt`, and the `eval` tool stays unrestricted by owner decision.
91
91
  - `ompSync.ts` runs `ompRemovals.ts syncOmpRemovals, retired-key inventory` right after the config merge, and that pass force-prunes retired kit-owned keys from `~/.omp/agent/config.yml` on every sync, without `--reconcile`. The pass is required because `mergeOmpConfig` is additive, so removing a key from the SoT alone never removes it from a deployed file. A key retired with a recorded value is pruned only while the deployed value still matches that value, so a user edit survives; a key retired outright, such as `providers.webSearchOrder`, is pruned at any value. The same pass removes each `bash.patterns` rule listed in `OMP_RETIRED_BASH_RULES`: the 39 file-removal rules retired in 0.20.9, the four `git push` prompts for a `$` or backtick word that 0.22.5 replaced with denies, and the 492 prompt rules retired in 0.22.8 (`RETIRED_PROMPT_MATCHES`). It compares with the merge's kit-rule normalization (approval case-folded, `match` whitespace collapsed), so a user rule with a different `match` or `approval` survives.
92
- - `cycleOrder` ends with `astra` as its fifth stop. `modelRoles.astra` is `openai-codex/gpt-6-astra:xhigh`, and `modelTags.astra` is visible. Astra and Fable fall back to each other through concrete selectors. `modelRoles.fable` is `anthropic/claude-fable-5-2:medium`, with visible `modelTags.fable`. The hidden `switch_fable` role uses the same Fable selector and keeps an empty fallback chain.
93
- - `modelRoles.task` is `anthropic/claude-opus-5-5:high`, and `retry.fallbackChains.task` falls back to `openai-codex/gpt-6.1-sol:high`. The kit relies on omp's default `task.enableEffort: false`: a caller `effort` outranks the role's level suffix, and `hi` maps to the top of the model ladder, so with the field on, `hi` spawns ran at `task.maxEffort` (`xhigh`) instead of the role level. With the field off, every spawn runs its role level, and `task.maxEffort` bounds only a per-call `model` with a `:level` suffix. By owner decision (0.22.22), the omp advisor is removed: the SoT has no `advisor` block, `advisor` role, or `advisor` fallback chain. The plan-lifecycle verification loop replaces a second model on every turn. A session that runs `/advisor on` takes its advisor model from the `slow` role. `smol`, `commit`, and `tiny` use `openai-codex/gpt-6-luna`. The kit-defined hidden `reviewer` role is `openai-codex/gpt-6.1-terra:high`. Its retry fallback is `anthropic/claude-opus-5-5:high`. All four reviewer entries in `task.agentModelOverrides` point to `@reviewer`. `task.agentServiceTierOverrides` sets the same four agents to `inherit`, so they take the live tier of the main session for each provider. The user set the effort `high` and chose to follow the fast toggle (#134); the review benchmark of #130 chose the model. `ompOverlay.ts ROLE_ORDER, FALLBACK_ORDER` list `reviewer`, so a free session also replaces it. omp bundles the `reviewer` and `security-reviewer` agents, and the `plan-lifecycle` plugin ships `code-reviewer` and `plan-reviewer`. The `SoT/toolchain.json` omp floor is 18.4.4, the oldest release the kit tested with `gpt-6.1-sol`. That release lists and serves the model through the `openai-codex` provider. `cli/docs/omp-models.md` records the checks.
94
- - The six Opus 5.5 roles (`default`, `slow`, `plan`, `designer`, `vision`, `task`) and the `smol`, `tiny`, `commit`, and `reviewer` retry chains use `anthropic/claude-opus-5-5` at the levels the role map records. `cli/docs/omp-models.md` carries the Artificial Analysis capture behind the role map, read 2026-09-29 at Intelligence Index v4.3.2 and Coding Agent Index v1.5 from the AA comparison-page metric tables. The `reviewer`, `fable`, and `switch_fable` rows are the exceptions: they come from v5.0.1 captures read 2026-11-05 (GPT-6.1 Terra) and 2026-11-12 (Fable 5.2), and they compare only with other v5.0.1 figures. Opus 5.5 max has the highest v4.3.2 index in that topic, at 58. AA has measured speed and latency for Opus 5.5 max and every GPT-6.1 Sol level; GPT-6 Luna medium is the only unmeasured speed and latency row.
92
+ - `cycleOrder` ends with `astra` as its fifth stop. `modelRoles.astra` is `openai-codex/gpt-6.2-astra:xhigh`, and `modelTags.astra` (`GPT-6.2 Astra`) is visible. Astra and Fable fall back to each other through concrete selectors: `retry.fallbackChains.astra` holds `anthropic/claude-fable-5-2:medium`, and `retry.fallbackChains.fable` holds `openai-codex/gpt-6.2-astra:xhigh`. `modelRoles.fable` is `anthropic/claude-fable-5-2:medium`, with visible `modelTags.fable`. The hidden `switch_fable` role uses the same Fable selector and keeps an empty fallback chain.
93
+ - `modelRoles.task` is `anthropic/claude-opus-5-5:high`, and `retry.fallbackChains.task` falls back to `openai-codex/gpt-6.2-sol:high`. The kit relies on omp's default `task.enableEffort: false`: a caller `effort` outranks the role's level suffix, and `hi` maps to the top of the model ladder, so with the field on, `hi` spawns ran at `task.maxEffort` (`xhigh`) instead of the role level. With the field off, every spawn runs its role level, and `task.maxEffort` bounds only a per-call `model` with a `:level` suffix. By owner decision (0.22.22), the omp advisor is removed: the SoT has no `advisor` block, `advisor` role, or `advisor` fallback chain. The plan-lifecycle verification loop replaces a second model on every turn. A session that runs `/advisor on` takes its advisor model from the `slow` role. `smol`, `commit`, and `tiny` use `openai-codex/gpt-6-luna`. The kit-defined hidden `reviewer` role is `openai-codex/gpt-6.1-terra:high`. Its retry fallback is `anthropic/claude-opus-5-5:high`. All four reviewer entries in `task.agentModelOverrides` point to `@reviewer`. `task.agentServiceTierOverrides` sets the same four agents to `inherit`, so they take the live tier of the main session for each provider. The user set the effort `high` and chose to follow the fast toggle (#134); the review benchmark of #130 chose the model. `ompOverlay.ts ROLE_ORDER, FALLBACK_ORDER` list `reviewer`, so a free session also replaces it. omp bundles the `reviewer` and `security-reviewer` agents, and the `plan-lifecycle` plugin ships `code-reviewer` and `plan-reviewer`. The `SoT/toolchain.json` omp floor is 18.22.2, the oldest release the kit tested with `gpt-6.2-sol` and `gpt-6.2-astra`. That release lists and serves both models through the `openai-codex` provider. The floor is doctor visibility only. Sync does not check the omp catalog, so an older omp can fuzzy-match a missing id to another model. `cli/docs/omp-models.md` records the checks.
94
+ - The six Opus 5.5 roles (`default`, `slow`, `plan`, `designer`, `vision`, `task`) and the `smol`, `tiny`, `commit`, and `reviewer` retry chains use `anthropic/claude-opus-5-5` at the levels the role map records. `cli/docs/omp-models.md` carries the Artificial Analysis capture behind the role map, read 2026-09-29 at Intelligence Index v4.3.2 and Coding Agent Index v1.5 from the AA comparison-page metric tables. The `astra`, `reviewer`, `fable`, and `switch_fable` rows are the exceptions: they come from v5.0.1 captures read 2026-11-18 (GPT-6.2 Astra), 2026-11-05 (GPT-6.1 Terra), and 2026-11-12 (Fable 5.2), and they compare only with other v5.0.1 figures. Opus 5.5 max has the highest v4.3.2 index in that topic, at 58. AA has measured speed and latency for every Opus 5.5 level and every GPT-6.2 Sol and GPT-6.2 Astra row in that topic. GPT-6 Luna medium is the only unmeasured speed and latency row.
95
95
  - `modelRoles.web` is `web/firecrawl` and `retry.fallbackChains.web` carries the explicit 20-entry provider order. The kit declares both keys because the legacy `providers.webSearchOrder` key is retired: omp expands it in memory into these two keys and then drops it, and never writes that expansion back to disk. An explicit chain replaces omp's built-in web order wholesale, so every entry left out is a provider omp never tries. The owner removed the seven entries that named older models (Gemini 2.5 Flash, Claude Haiku 4.5, GPT-5.6, GPT-5.5, and Grok 4.5); keep every other provider.
96
- - `SoT/.omp/models.yml` declares Astra's full `low, medium, high, xhigh, max` ladder with `defaultLevel: xhigh` as the worked provider ladder-override example. The temporary `anthropic.modelOverrides.claude-opus-5-5` block is gone, because the shared catalog now publishes the same limits, ladder, and prices. The removal switches omp from the block's `thinking.mode: effort` (a `budget_tokens` value per level from `ANTHROPIC_THINKING`, no effort value) to the catalog's `anthropic-adaptive` (adaptive thinking with the role's effort level); every kit Opus selector names its level, so the block's bare-selector `defaultLevel: high` is not needed. `ompRemovals.ts syncOmpModelRemovals, retired-block inventory` prunes it from `~/.omp/agent/models.yml` after the models merge, but only while the deployed block still equals the shipped one. `ompYaml.ts mergeOmpModels` preserves deployed-only keys in `~/.omp/agent/models.yml`, because a user file may carry provider credentials. Whole-file replacement is wrong.
96
+ - `SoT/.omp/models.yml` declares Astra's full `low, medium, high, xhigh, max` ladder with `defaultLevel: xhigh` as the worked provider ladder-override example. The override is keyed `gpt-6.2-astra`, the model id of the `astra` role. `ompRemovals.ts syncOmpModelRemovals` prunes the old `gpt-6-astra` block from `~/.omp/agent/models.yml` while it equals the shipped block. The temporary `anthropic.modelOverrides.claude-opus-5-5` block is gone, because the shared catalog now publishes the same limits, ladder, and prices. The removal switches omp from the block's `thinking.mode: effort` (a `budget_tokens` value per level from `ANTHROPIC_THINKING`, no effort value) to the catalog's `anthropic-adaptive` (adaptive thinking with the role's effort level); every kit Opus selector names its level, so the block's bare-selector `defaultLevel: high` is not needed. `ompRemovals.ts syncOmpModelRemovals, retired-block inventory` prunes it from `~/.omp/agent/models.yml` after the models merge, but only while the deployed block still equals the shipped one. `ompYaml.ts mergeOmpModels` preserves deployed-only keys in `~/.omp/agent/models.yml`, because a user file may carry provider credentials. Whole-file replacement is wrong.
97
97
  - `SoT/.omp/AGENTS.md` carries the rule `Please remove all mannered prose.` Anthropic's Fable 5.1 prompting guide documents mannered prose as a Fable 5.1 behavior and gives that sentence as its short-version fix: https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/prompting-claude-fable-5-1
98
- - `cli/docs/omp-models.md` (topic `omp-models`) records the role map rationale and the Artificial Analysis snapshot behind it. Model choices change with published benchmarks, so update that topic in the same commit as a role change.
98
+ - `cli/docs/omp-models.md` (topic `omp-models`) records the role map rationale and the Artificial Analysis snapshot behind it. Model choices change with published benchmarks, so update that topic in the same commit as a role change. For each model that an omp role or retry chain uses, the topic keeps AA rows at `high` and `xhigh` at least, with the index version and the capture date.
99
99
  - `SoT/.omp/config.yml` carries only keys whose value differs from the omp default. When a kit value equals the omp default, remove the key from the SoT. Add it with its shipped value to `ompRemovals.ts OMP_RETIRED_VALUES`, because `ompYaml.ts mergeOmpConfig` is additive and only that pass prunes the deployed copy. For example, the kit ships no `compaction.thresholdTokens`, and sync prunes a deployed `-1` and the former `231200` pin. `cli/docs/omp-context.md` (topic `omp-context`) carries the reserve-based derivation and the measured evidence, so update that topic in the same commit as any compaction-setting change.
100
- - Sync materializes the kit omp plugin `docks-kit-omp` from `SoT/.omp/plugin/` to `~/.docks-kit/omp-plugin` (`ompKitPlugin.ts syncKitPlugin`). It sets directories to 0700, files to 0600, and the `package.json` `version` to the kit version, and it removes stale files. It keeps no backups, because the directory is kit-owned. `ompPlugins.ts syncPlugins, kit plugin link` then runs `omp plugin link <dir>`. It skips the link when the `omp plugin list --json` npm row `docks-kit-omp` already has a `path` that resolves to that directory. `--skip-plugin-refresh` does not skip the link. The `hygiene.ts` extension registers `scratch_new`, `scratch_remove`, and `tidy_report`. It adds a reminder after each bash command that created a resource, and it checks at session stop that those resources are gone. At a main-session start, it claims `~/.cache/docks-kit/tidy.stamp.claim` and runs `docks-kit tidy --apply` in the background at most once per 12 hours. Every extension subprocess runs without a shell. The `pluginUpgrades.ts` extension reads the omp plugin upgrade log. In a tool result of each main session, it adds one reminder for the logged upgrades that the session did not get before. It names only the upgrades that finished after its omp process started. The reminder tells the agent to ask the user to run `/reload-plugins`. It then tells the agent to read each plugin skill again through `skill://` before it runs a skill script. It asks for a restart instead when an upgraded plugin ships hooks, extension modules, or non-MCP tools. omp imports extensions only at startup. An omp process that started before a sync deployed `pluginUpgrades.ts` did not load it, so its sessions get no reminder. The plugin also ships the `workspace-hygiene` skill and the `/cleanup` command. The plugin, the deployed rules, and the sweep call `docks-kit scratch` and `docks-kit tidy`. So a checkout sync must not run before the global `docks-kit` carries those commands.
100
+ - Sync materializes the kit omp plugin `docks-kit-omp` from `SoT/.omp/plugin/` to `~/.docks-kit/omp-plugin` (`ompKitPlugin.ts syncKitPlugin`). It sets directories to 0700, files to 0600, and the `package.json` `version` to the kit version, and it removes stale files. It keeps no backups, because the directory is kit-owned. `ompPlugins.ts syncPlugins, kit plugin link` then runs `omp plugin link <dir>`. It skips the link when the `omp plugin list --json` npm row `docks-kit-omp` already has a `path` that resolves to that directory. `--skip-plugin-refresh` does not skip the link. The `hygiene.ts` extension registers `scratch_new`, `scratch_remove`, and `tidy_report`. It adds a reminder after each bash command that created a resource, and it checks at session stop that those resources are gone. At a main-session start, it claims `~/.cache/docks-kit/tidy.stamp.claim` and runs `docks-kit tidy --apply` in the background at most once per 12 hours. Every extension subprocess runs without a shell. The `pluginUpgrades.ts` extension reads the omp plugin upgrade log. In a tool result of each main session, it adds one reminder for the logged upgrades that the session did not get before. It names only the upgrades that finished after its omp process started. The reminder tells the agent to ask the user to run `/reload-plugins`. It then tells the agent to read each plugin skill again through `skill://` before it runs a skill script. It asks for a restart instead when an upgraded plugin ships hooks, extension modules, or non-MCP tools. omp imports extensions only at startup. An omp process that started before a sync deployed `pluginUpgrades.ts` did not load it, so its sessions get no reminder. The plugin also ships the `workspace-hygiene` skill, the `/cleanup` command, and six omp Time Traveling Stream Rules (TTSR, trigger rules) in `rules/`: `file-removal.md`, `git-safety.md`, `temp-and-background.md`, `worktree-location.md`, `literal-paths.md`, and `remote-cleanup.md`. omp's `omp-plugins` provider loads `rules/*.md` from the linked plugin root (`omp://plugin-manager-installer-plumbing.md`). Each rule has a regex `condition` scoped to `tool:bash, tool:eval` that aims at the wrong form; eval is in scope because eval code can run a shell command through `sh()`. A condition reads text, not a parsed command, so it has known false positives: a quoted command word (`echo "rm x"`, or a commit message that starts with `rm`) and eval code after the closing quote of a `sh()` string. `worktree-location.md` fires on each `git worktree add`, also at the correct path, because a condition cannot read `$XDG_DATA_HOME` or the home directory. omp aborts the matching tool call before it runs, injects the rule, and retries (`omp://ttsr-injection-lifecycle.md`). Each rule has `repeatMode: always`: it fires at each later wrong form, but not on the retry that its own reminder starts. The rules reach subagents and cost no prompt tokens until they fire. They teach the correct form; the `bash.patterns` denies still enforce. omp tests each condition on the partial argument buffer after each streamed delta, so a condition must not match a prefix of a correct command: it needs a delimiter after each token it checks, never the end anchor `$`. The buffer is the argument JSON text, the intent field `i` can come before or after `command`, and a provider can put JSON white space around the colon (`{"command": "ls"}`). So each condition starts its match inside the `command` or `code` string with the lookbehind `(?<="(?:command|code)"\s*:\s*"(?:\\.|[^"\\])*)`, each scan stops at an unescaped `"` and at a newline (`\n`), a tab (`\t` in the buffer) counts as shell white space, and a scan that looks for an option stops at a ` -- ` word. `cli/test/unit/ompTriggerRules.test.ts` pins every condition through bash and through Python and JavaScript eval `sh()` calls, in the compact and the spaced JSON layouts, with some rows in the pretty layout and trigger words in the intent before and after the command. It checks that each condition has a wrong row that fires it and that each rule has `repeatMode: always`. `cli/docs/command-guards.md` (topic `command-guards`) lists the rules and the false positives. The plugin, the deployed rules, and the sweep call `docks-kit scratch` and `docks-kit tidy`. So a checkout sync must not run before the global `docks-kit` carries those commands.
101
101
  - `task.agentIdleTtlMs` is `120000`, so omp parks an idle subagent after 2 minutes instead of 7. `browser.idleCloseSec` is `600`.
102
- - By owner decision (0.21.0), resources on the workstation are always development, so `docks-kit tidy` sweeps them without a question. The omp agent prompt, the `workspace-hygiene` skill, `/cleanup`, and the extension reminders limit the cleanup of a remote resource (reached through SSH or a provider CLI such as `wrangler`, `vercel`, `kubectl`, or a remote Docker host) to targets named local, development, dev, test, homolog, homologation, staging, or preview, the plan-lifecycle non-production definition. A production cleanup needs an `ask` immediately before it, must be rare, and a `yolo` invocation does not answer that ask. Prompt rules cannot force an agent to ask, so this rule has no `bash.patterns` guard.
102
+ - By owner decision (0.21.0), resources on the workstation are always development, so `docks-kit tidy` sweeps them without a question. The omp `remote-cleanup.md` trigger rule, the `workspace-hygiene` skill, `/cleanup`, and the extension reminders limit the cleanup of a remote resource (reached through SSH, a database client, or a provider CLI such as `wrangler`, `vercel`, `kubectl`, `aws`, or a remote Docker host) to targets named local, development, dev, test, homolog, homologation, staging, or preview, the plan-lifecycle non-production definition. A production cleanup needs an `ask` immediately before it, must be rare, and a `yolo` invocation does not answer that ask. Prompt rules cannot force an agent to ask, so this rule has no `bash.patterns` guard.
103
103
  - The omp CLI is upstream-owned. It updates through `omp update`, or through the external manager that owns the install, such as Tern or Nix. `omp update` refuses to run on an install that an external manager owns. Sync never installs or upgrades the CLI.
104
104
  - omp plugin upgrade flow, the `pi-intercom` install, and the `docks-kit omp` free-model launcher: `cli/AGENTS.md`.
105
105
 
@@ -22,10 +22,16 @@ subagents. This includes any `rm` that omp does not deny, `git clean`,
22
22
  `git reset --hard`, a force push that no deny matches, and a git command
23
23
  that skips the repository's hooks. A prompt rule that the user adds to
24
24
  `~/.omp/agent/config.yml` survives sync and still asks; a subagent rejects
25
- it. The deployed omp agent prompt forbids `--force`, `-f`, and their
26
- abbreviations with `git push`, and it forbids each hook skip. It tells the
27
- agent to use `ask` before a force push, and to ask the user before
28
- `git clean` or `git reset --hard`.
25
+ it. The kit omp plugin trigger rule `git-safety.md` forbids these options with
26
+ `git push` and `git rm`: `-f`, a flag cluster that holds `f`, and a long
27
+ option that starts with `--f`. `git push --follow-tags` is correct. The rule
28
+ also forbids `git push origin +<ref>`, `git push --mirror`,
29
+ `git push --prune`, and each hook skip. It tells the agent to ask before a
30
+ bulk discard in the main session, for example `git clean` without `-n`,
31
+ `git reset --hard`, `git checkout .`, or `git stash clear`.
32
+ A subagent does not run a bulk discard and names it in its report. The omp
33
+ agent prompt lists a force push as a step that needs `ask`. The section
34
+ "omp trigger rules" below describes these rules.
29
35
 
30
36
  The policy sorts shell commands into three tiers. omp uses two of them: deny
31
37
  and allow.
@@ -61,7 +67,7 @@ that `eval` starts.
61
67
  | A `$` or a backtick anywhere after `git push` in the same command | deny, also after a shell keyword, a wrapper, or a path before `git`: see below | ask for a word that starts with `$` or a backtick | ask (every `git push` asks) |
62
68
  | A git command that skips the repository's hooks: see below | run | no rule; the classifier decides | ask for each `git push`, `git merge`, and `git rebase` that starts the command; no rule for `git commit`, `git am`, or `git pull` |
63
69
  | `git rm -- <path>`, `git worktree remove -- <path>` | run | ask | ask |
64
- | `git worktree remove --force -- <path>` | run; the omp agent prompt permits it once, after the commits are on the remote | ask | ask |
70
+ | `git worktree remove --force -- <path>` | run; the omp `file-removal.md` trigger rule fires on each `--force`, `-f`, or `--f` form and permits it once, after the commits are on the remote | ask | ask |
65
71
  | `docks-kit scratch remove <path>` | run; the command removes only a path inside the scratch root, the system temporary directory, or `/var/tmp` | no rule | no rule |
66
72
  | Any other non-recursive `git rm` or `git worktree remove` | run | ask | ask |
67
73
  | `gio trash -- <path>` | run | allow | ask |
@@ -97,19 +103,20 @@ branch name.
97
103
  omp therefore denies a `$` or a backtick anywhere after `git push` in the same
98
104
  command. The agent reads a refusal such as
99
105
  `Blocked by bash pattern: *git push *$*`, writes the ref literally, and runs
100
- the push without a question. The omp agent prompt names the literal forms:
101
- `git push -u origin HEAD` for the current branch, and
106
+ the push without a question. The omp `git-safety.md` trigger rule names the
107
+ literal forms: `git push -u origin HEAD` for the current branch, and
102
108
  `git push origin --delete <branch-a> <branch-b>` for several branches. The two
103
- omp rules start with `*`. So a shell keyword such as `do` in a loop, a wrapper
104
- such as `env`, or a path such as `/usr/bin/git` does not hide the push.
109
+ omp deny rules start with `*`. So a shell keyword such as `do` in a loop, a
110
+ wrapper such as `env`, or a path such as `/usr/bin/git` does not hide the
111
+ push.
105
112
 
106
113
  ## Commands that skip the repository's hooks
107
114
 
108
115
  A git command can skip the repository's hooks, for example with
109
116
  `--no-verify`, `git commit -n`, `git am -n`, or a `core.hooksPath` change.
110
- The kit ships no omp rule for these commands, so omp runs them without a
111
- question. The deployed omp agent prompt forbids each hook skip. Claude Code
112
- has no rule either, so its classifier decides. Codex asks for each
117
+ The kit ships no omp `bash.patterns` rule for these commands, so omp does not
118
+ block them. The omp `git-safety.md` trigger rule forbids each hook skip.
119
+ Claude Code has no rule either, so its classifier decides. Codex asks for each
113
120
  `git push`, `git merge`, and `git rebase` that starts the command. Codex has
114
121
  no rule for `git commit`, `git am`, or `git pull`.
115
122
 
@@ -152,8 +159,9 @@ no rule for `git commit`, `git am`, or `git pull`.
152
159
  7. Write literal absolute paths. Never use `~` or `$HOME`, and never set or
153
160
  export `HOME`.
154
161
 
155
- The deployed agent prompts (`~/.omp/agent/AGENTS.md`, `~/.claude/CLAUDE.md`,
156
- and the Codex `AGENTS.md`) carry the same rules.
162
+ In omp, the kit plugin trigger rules `file-removal.md` and `literal-paths.md`
163
+ carry these rules. `~/.claude/CLAUDE.md` and the Codex `AGENTS.md` carry the
164
+ same rules.
157
165
 
158
166
  ### What the git removals keep
159
167
 
@@ -167,6 +175,87 @@ and the Codex `AGENTS.md`) carry the same rules.
167
175
  short-option cluster, such as `-rff`. After `--`, git reads every argument as
168
176
  a path. So the `--` form is the only one that cannot force.
169
177
 
178
+ ## omp trigger rules
179
+
180
+ omp loads `~/.omp/agent/AGENTS.md` only into main sessions and advisors,
181
+ never into task subagents (`omp://context-files.md`). So the kit puts the omp
182
+ command rules into six Time Traveling Stream Rules (TTSR, trigger rules). The
183
+ kit plugin `docks-kit-omp` ships them in `SoT/.omp/plugin/rules/`. Sync
184
+ materializes the plugin to `~/.docks-kit/omp-plugin` and links it. omp's
185
+ `omp-plugins` provider loads `rules/*.md` from the linked plugin root. Each
186
+ rule has the scope `tool:bash, tool:eval`, because eval code can run a shell
187
+ command through `sh()`.
188
+
189
+ | Rule | Scope | Correct form that the rule teaches |
190
+ | --- | --- | --- |
191
+ | `file-removal.md` | `tool:bash`, `tool:eval` | `scratch_remove` or `docks-kit scratch remove <path>` for a temporary path; `gio trash -- <path>` or `trash <path>` for another path; no `rm`, `unlink`, or `shred`, also after a `VAR=value` prefix; `git rm -- <path>`; for a folder, `git rm -- '<dir>/*'` or `git rm --cached -- '<dir>/*'`, because omp denies `git rm -r`; `git worktree remove -- <path>`, and `--force` only after the commit and status checks; no deletion through `eval` |
192
+ | `git-safety.md` | `tool:bash`, `tool:eval` | no hook skip, also through an abbreviation such as `--no-veri`; no `-f`, `f` cluster, or `--f` long option with `git push` or `git rm` (`--follow-tags` is correct); no `+<ref>` with `git push` (`git push origin +<ref>` is a force push); no `git push --mirror` or `--prune`; a literal push ref; `git push` as its own command, not chained with `&&`, `\|\|`, or `;`; `git push -u origin HEAD`; `git push origin --delete <branch-a> <branch-b>`; `git branch -d <branch>`, not `-D` or `-d --force`; `git restore -- <path>`; an ask before `git clean` without `-n` or `--dry-run` before ` -- `, `git reset --hard`, `git checkout .`, `git restore .` (`git restore --staged .` only unstages and is correct), `git stash drop`, or `git stash clear`. The git global options `-C`, `-c`, `-p`, `-P`, and `--<option>` do not hide the subcommand. A word after ` -- ` is a path, not an option |
193
+ | `temp-and-background.md` | `tool:bash`, `tool:eval` | `scratch_new` or `docks-kit scratch new <label>` instead of `/tmp`; a server or watcher only as a named bash service or an async job; no `nohup`, `setsid`, `disown`, `tmux`, or `screen`, also after a `VAR=value` prefix; no global install (`npm`, `bun`, `pnpm`, `yarn`, `pipx`, `cargo`, `go install`, `gem install`, `uv tool install`), scheduled job, or enabled service; no Docker cleanup of a resource that the agent did not create |
194
+ | `worktree-location.md` | `tool:bash`, `tool:eval` | a worktree only at the literal absolute path `<data>/agent-worktrees/<repository>/<name>`, with `agent-worktrees` as a whole directory name. The rule fires on each `git worktree add`, also at the correct path, because a condition cannot read `$XDG_DATA_HOME` or the home directory of the user. The retry after the reminder runs |
195
+ | `literal-paths.md` | `tool:bash`, `tool:eval` | a literal absolute path; no `~` or `$HOME`; no `HOME` assignment |
196
+ | `remote-cleanup.md` | `tool:bash`, `tool:eval` | a remote cleanup only on a non-production target; an `ask` immediately before a production cleanup, which must be rare. Remote means SSH, `rsync --delete` to a `host:` target, `rclone` (`delete` verbs, and `sync` to a `remote:` target), `gh` (`gh <noun> delete`, `gh <noun> <x>-delete`, and `gh api -X DELETE`), a remote `docker` or `podman` (also `docker -c <context>`), a database client (`DROP TABLE`, `DATABASE`, or `SCHEMA` in `psql`, `mysql`, `sqlite3`, or `mongosh`; `dropdb`; `redis-cli FLUSHALL` or `FLUSHDB`), or a provider CLI with a cleanup verb, such as `aws s3 rb` or `neonctl branches delete` |
197
+
198
+ Each rule has a regex `condition` that aims at the wrong form. When the
199
+ model streams a matching tool call, omp aborts the call before it runs,
200
+ injects the rule body, and retries (`omp://ttsr-injection-lifecycle.md`). A
201
+ rule costs no prompt tokens until it fires, and it reaches subagents too. The
202
+ trigger rules teach the correct form. The `bash.patterns` denies do not
203
+ change, and they still enforce. Each rule has `repeatMode: always`, so it
204
+ fires at each later wrong form in the session too. The retry that its own
205
+ reminder starts does not fire it again, so a correct command still runs.
206
+
207
+ After a session resume, omp restores each injected rule at message count
208
+ zero. So a rule that fired before the resume does not fire in the first turn
209
+ after the resume (section 7 of
210
+ [ttsr-injection-lifecycle.md](https://github.com/can1357/oh-my-pi/blob/main/docs/ttsr-injection-lifecycle.md)).
211
+ No repeat mode changes this. The `bash.patterns` denies and the main-session
212
+ `ask` rules still apply in that turn.
213
+
214
+ A condition reads text, not a parsed shell command. So these correct calls
215
+ also fire a rule:
216
+
217
+ - A quoted command word, such as `echo "rm x"` or a commit message that
218
+ starts with `rm`, fires `file-removal.md`.
219
+ - In eval code, a scan can continue past the closing quote of the `sh()`
220
+ string to the end of the statement. For example,
221
+ `print(sh("kubectl get pods"), "pods to delete")` fires `remote-cleanup.md`.
222
+
223
+ omp tests each condition on the partial tool-call buffer after each streamed
224
+ delta, and again on the full arguments at the end of the call
225
+ (`omp://ttsr-injection-lifecycle.md`). So a condition must not match a prefix
226
+ of a correct command. Each condition must require a delimiter after the token
227
+ that it checks, and no condition uses the end anchor `$` as that delimiter.
228
+ A negative lookahead, such as `(?!ol)` in `--f(?!ol)`, passes at the end of
229
+ a partial buffer. So after a negative lookahead, the condition must read the
230
+ token to its delimiter. `git clean` and `git restore .` read the full
231
+ command before they decide, because `-n` or `--staged` can come last.
232
+
233
+ The buffer is the argument JSON text, and omp's schema puts the intent field
234
+ `i` before `command`. A call can also put `i` after `command`, and a provider
235
+ can stream JSON with white space, such as `{"command": "ls"}`. Each condition
236
+ has the lookbehind `(?<="(?:command|code)"\s*:\s*"(?:\\.|[^"\\])*)` before
237
+ its first command token, so a match starts only inside the `command` or
238
+ `code` string, with or without JSON white space around the colon. omp 18.21.0
239
+ accepts this variable-length lookbehind. A scan reads an escape pair, such as
240
+ `\"` or `\\`, as one unit. It stops at an unescaped `"`, which ends the
241
+ command string, and at a newline (`\n`), which ends the command line. So the
242
+ intent text and the next line do not fire a rule. A tab in a command is `\t`
243
+ in the buffer, so each condition reads `\t` as shell white space, as it reads
244
+ a space. A scan that looks for an option also stops at a ` -- ` word, also
245
+ with a tab, because each later word is a path. So `git clean -fd -- -n`
246
+ fires, and `git rm -- -f` stays quiet. In the `git push` conditions for `$`,
247
+ a backtick, and a chain, a scan also stops at `)` and at a backtick, which end
248
+ a JavaScript `` sh`…` `` or `sh("…")` call. A backtick counts as a command
249
+ substitution only when a word follows it.
250
+ `cli/test/unit/ompTriggerRules.test.ts` runs each shell row through bash,
251
+ Python `sh("…")`, JavaScript `` sh`…` ``, and JavaScript `sh("…");`. It runs
252
+ every row in the compact JSON layout (`{"command":"ls"}`) and in the spaced
253
+ layout (`{"command": "ls"}`), and some rows in the pretty layout, with a
254
+ newline and an indent. Other rows put trigger words in the intent field,
255
+ before or after the command, or put a tab in the command. The test checks
256
+ that each correct form stays quiet on each prefix, that each condition has a
257
+ wrong row that fires it, and that each rule has `repeatMode: always`.
258
+
170
259
  ## Why omp denies instead of asking
171
260
 
172
261
  omp has a built-in check for critical commands. For example, it flags:
@@ -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. |