stacktrace-cli 0.11.0__tar.gz → 0.13.0__tar.gz

{stacktrace_cli-0.11.0 → stacktrace_cli-0.13.0}/.github/workflows/ci.yml

@@ -5,6 +5,12 @@ on:
     branches: [main]
   pull_request:
 
+# A new push to a PR cancels the run for the push it replaces, as in Fleet.
+# Runs on main are never cancelled, so every merged commit is checked.
+concurrency:
+  group: ci-${{ github.workflow }}-${{ github.ref }}
+  cancel-in-progress: ${{ github.event_name == 'pull_request' }}
+
 jobs:
   lint-and-test:
     runs-on: ubuntu-latest

{stacktrace_cli-0.11.0 → stacktrace_cli-0.13.0}/CLAUDE.md

@@ -240,13 +240,23 @@ the alternative we rejected? If yes, write the ADR.
 full ADR. The cost of one Read is much smaller than the cost of
 re-deriving or re-litigating a decision.
 
-**Supersede, never edit.** Accepted ADRs are immutable. If a later
-decision contradicts an existing ADR, write a NEW ADR with
+**Supersede or amend, never edit.** Accepted ADRs are immutable. If a later
+decision contradicts an existing ADR, write a NEW ADR that supersedes or
+amends it, as the next rule decides. For a supersession, put
 `supersedes: NNNN` in its frontmatter, and update the old one's
 frontmatter to `status: superseded` + `superseded-by: NNNN`. Old PRs
 need to remain readable against the rules in effect at the time —
 silently editing an accepted ADR breaks that contract.
 
+**Amend or supersede: one test.** Ask if any decision in the old ADR still
+binds, whether or not the ADR numbers its decisions. If one does, amend: put
+`amends: NNNN` on the new ADR and `amended-by: NNNN` on the old one, and both
+stay active. If none does, supersede. Only this test decides. The size of
+the change, the importance of the replaced clause, and the old ADR's "When
+to revisit" text do not change the result. `docs/adrs/INDEX.md` gives the
+worked example. A reviewer reports the choice only when the finding names
+every decision in the old ADR and shows that none still binds.
+
 ## Commits and PRs
 
 - Frequent commits, one logical change per commit.
@@ -264,12 +274,55 @@ silently editing an accepted ADR breaks that contract.
   the user explicitly says "merge locally," "keep as branch," or
   "discard."
 
+## Writing documents
+
+Write all Markdown documents in ASD-STE100 Simplified Technical English.
+This includes specs, ADRs, READMEs, plans and release notes. The rule
+applies to every author, human or agent.
+
+- Write one idea in each sentence.
+- Use a maximum of 20 words in an instruction and 25 words in a
+  description.
+- Use a maximum of six sentences in a paragraph.
+- Use the active voice and the present tense.
+- Start an instruction with a verb, for example "Run the tests."
+- Use "a", "an" and "the". Do not remove them to make the text shorter.
+- Use one term for one thing, and use it the same way in all of the
+  document.
+- Use simple, common words. Do not use filler, such as "in order to" or
+  "utilize".
+- Do not use em-dash asides. Write a new sentence.
+- Do not change technical names, commands or code. Code blocks, tables and
+  quoted text do not have to obey these rules.
+
+Apply these rules to the text that you write or change. Do not rewrite an
+existing document only to change its style. Do not edit an accepted ADR to
+change its style.
+
+These rules are for authors. They are not a code review topic. Reviewers
+do not report style findings.
+
 ## Code Review Rules
 
+**The goal of review is to converge.** The reviewer and the fixer both work
+toward a PR that can merge in few rounds. Each round costs a full CI run and
+a review, and a finding on text that was already reviewed starts another
+round. Converging does not mean letting a serious defect through. The first
+review catches the glaring issues, and later reviews check only what changed.
+
 ### Reviewer
 
-- Review the full PR against its base branch. Report all qualifying
-  findings together; do not deliberately reserve findings for later rounds.
+- **The first review of a PR is the complete one.** Review the full PR
+  against its base branch. Report every qualifying finding together. Do not
+  reserve findings for later rounds.
+- **A later review covers only what changed since the previous reviewed
+  head.** That is the PR's new commits, plus existing code whose behavior
+  those commits change, such as a caller or dependency of changed code.
+  Changes that a merge or rebase brings in from the base branch are not part
+  of it.
+- **Do not report a defect that existed at the previous reviewed head**,
+  unless the new commits change its behavior. Exception: report a P0 or P1
+  even if it existed before, and label it a late discovery.
 - Report concrete, actionable defects with supported failure scenarios.
   Respect explicit scope decisions and accepted tradeoffs. Do not present
   speculative hardening or optional improvements as correctness defects.
@@ -280,12 +333,16 @@ silently editing an accepted ADR breaks that contract.
   - P3: low-priority suggestion.
 - Report only P0 and P1 findings. Do not post P2 or P3 findings, inline or
   in the review summary. Do not inflate priority to get a finding reported.
-- On subsequent reviews, verify earlier fixes and inspect their effects on
-  callers and dependencies. Older code within the PR remains reviewable.
+- A missing detail or an optional improvement is never P1. This applies to
+  code and to documentation. A P1 is something wrong: a defect, or a false
+  statement.
+- On later reviews, verify that the earlier fixes work. Check their effects on
+  callers and dependencies.
 - When review history supports it, identify a finding as:
   - Regression: introduced since the previous reviewed head.
-  - Late discovery: present at a previously reviewed head but not reported.
   - Unresolved: a previously reported defect remains.
+  - Late discovery: present at a previously reviewed head but not reported.
+    Only a P0 or P1 may be a late discovery.
   If the history is unavailable or ambiguous, say so rather than guessing.
 - Deduplicate by underlying defect and remedy, not by title. Refer to an
   existing thread for an unresolved defect instead of opening another one.
@@ -301,6 +358,13 @@ silently editing an accepted ADR breaks that contract.
   If its reasoning or priority is wrong, explain why rather than applying
   it solely because a reviewer requested it. Surface unresolved disputes
   for human judgment.
+- Decline a late discovery that is not P0 or P1. If the defect existed at the
+  previous reviewed head and the new commits do not change its behavior,
+  reply that the review rules exclude it, and do not fix it. A declined
+  finding does not count as a fix round.
+- Keep each fix to what its finding needs. Do not rewrite or extend nearby
+  code or text. A change beyond the finding gives the next review new
+  material to report, and the PR does not converge.
 - Fix the underlying invariant across relevant call sites. Test the
   failure class rather than only the reported example.
 - Preserve accepted fixes and regression tests when reviewing a rebased PR or
@@ -310,9 +374,18 @@ silently editing an accepted ADR breaks that contract.
 - Batch related fixes into one tested update before requesting re-review.
   Avoid duplicate review requests for the same head.
 - An automated fixer pushes ordinary commits to the existing PR branch and
-  runs every required gate. It does not bypass gates, merge, rebase, or
-  force-push, or change workflows, permissions, credentials, or branch
-  protection. Report an environment failure or required rebase as a blocker.
+  runs every required gate. It does not bypass gates, merge, or change
+  workflows, permissions, credentials, or branch protection. It force-pushes
+  only after a rebase, as the next rule says. Report an environment failure
+  as a blocker.
+- When a PR branch conflicts with its base branch, the automated fixer
+  rebases it onto the current base branch. The base branch is normally
+  `main`. It resolves the conflicts, runs every required gate, and pushes
+  with `git push --force-with-lease`. It does not merge the base branch into
+  the PR branch. PR branches are rarely shared, so rewritten history is
+  acceptable. Report a conflict as a blocker when both sides change the same
+  logic and each resolution loses behavior. A rebase with no other change is
+  not a fix round.
 - After pushing fixes, wait for CI and a completed review of the current
   head. Silence, an older review, or a running review is not clearance.
 - Stop and ask for human input when the same failure repeats without progress.

{stacktrace_cli-0.11.0 → stacktrace_cli-0.13.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.5
 Name: stacktrace-cli
-Version: 0.11.0
+Version: 0.13.0
 Summary: CLI for Stacktrace — Detection and Response platform for AI Agents.
 Project-URL: Homepage, https://stacktrace.ai
 Author-email: "Stacktrace AI, Inc" <founders@stacktrace.ai>
@@ -20,8 +20,8 @@ Classifier: Topic :: Security
 Requires-Python: >=3.11
 Requires-Dist: click>=8.1
 Requires-Dist: httpx<1.0.dev0,>=0.28.1
-Requires-Dist: openaca==0.10.1
-Requires-Dist: openaidr==0.6.3
+Requires-Dist: openaca==0.11.0
+Requires-Dist: openaidr==0.7.0
 Requires-Dist: posthog<8,>=7.0
 Requires-Dist: pydantic<3,>=2.10
 Description-Content-Type: text/markdown
@@ -52,7 +52,7 @@ Requires Python 3.11 or newer.
 
 ```console
 $ stacktrace --version
-stacktrace 0.11.0 (openaca 0.10.1)
+stacktrace 0.13.0
 ```
 
 ## Python package namespace
@@ -107,9 +107,14 @@ stacktrace webhook disable
 ```
 
 `configure` prompts for an optional HMAC-SHA256 signing secret without echoing
-it. The endpoint receives a narrow JSON document containing the event ID, rule,
-severity, confidence and session identity—not transcript content, finding prose
-or evidence. Delivery is at least once, so receivers should deduplicate on
+it. The endpoint receives a narrow JSON document: the event ID, detection time,
+rule, severity, confidence and session identity, plus, when the finding has
+them, the rule's catalogue title, the component it concerns (type and name) and
+the kinds of evidence it cites. For `stacktrace-destructive-action` the
+evidence kind is `rm -rf` and the paths it deleted outside the workspace,
+redacted so an absolute path is never sent (ADR-0073, ADR-0084). Transcript
+content, remediation text and evidence detail are not sent. Delivery is at
+least once, so receivers should deduplicate on
 `event_id`. The URL and secret are stored in a mode-0600 host configuration
 file. HTTP is accepted only for loopback development endpoints.
 
@@ -250,6 +255,9 @@ Beta, and under active development.
 Devin sessions alike (Devin CLI and Devin Desktop's local agent). Session collection is
 [OpenAIDR](https://github.com/open-agent-security/openaidr)'s, and further agent
 kinds arrive with it. The daemon observes every kind it reads, sub-agents included.
+[Agent support](docs/agent-support.md) lists what works for each agent, rule by
+rule, and what cannot. Codex is next
+([spec](docs/specs/codex-agent-kind.md)).
 
 ## Built on
 

{stacktrace_cli-0.11.0 → stacktrace_cli-0.13.0}/README.md

@@ -24,7 +24,7 @@ Requires Python 3.11 or newer.
 
 ```console
 $ stacktrace --version
-stacktrace 0.11.0 (openaca 0.10.1)
+stacktrace 0.13.0
 ```
 
 ## Python package namespace
@@ -79,9 +79,14 @@ stacktrace webhook disable
 ```
 
 `configure` prompts for an optional HMAC-SHA256 signing secret without echoing
-it. The endpoint receives a narrow JSON document containing the event ID, rule,
-severity, confidence and session identity—not transcript content, finding prose
-or evidence. Delivery is at least once, so receivers should deduplicate on
+it. The endpoint receives a narrow JSON document: the event ID, detection time,
+rule, severity, confidence and session identity, plus, when the finding has
+them, the rule's catalogue title, the component it concerns (type and name) and
+the kinds of evidence it cites. For `stacktrace-destructive-action` the
+evidence kind is `rm -rf` and the paths it deleted outside the workspace,
+redacted so an absolute path is never sent (ADR-0073, ADR-0084). Transcript
+content, remediation text and evidence detail are not sent. Delivery is at
+least once, so receivers should deduplicate on
 `event_id`. The URL and secret are stored in a mode-0600 host configuration
 file. HTTP is accepted only for loopback development endpoints.
 
@@ -222,6 +227,9 @@ Beta, and under active development.
 Devin sessions alike (Devin CLI and Devin Desktop's local agent). Session collection is
 [OpenAIDR](https://github.com/open-agent-security/openaidr)'s, and further agent
 kinds arrive with it. The daemon observes every kind it reads, sub-agents included.
+[Agent support](docs/agent-support.md) lists what works for each agent, rule by
+rule, and what cannot. Codex is next
+([spec](docs/specs/codex-agent-kind.md)).
 
 ## Built on
 

stacktrace_cli-0.13.0/SECURITY.md

@@ -0,0 +1,41 @@
+# Security policy
+
+## Reporting a vulnerability
+
+Email **security@stacktrace.ai**. Please do not open a public issue, post in a
+public channel, or disclose details before we have released a fix.
+
+A useful report says:
+
+- the `stacktrace --version` output and your operating system;
+- what you did, what happened, and what you expected;
+- whether the issue can expose session content, credentials or file paths, or
+  send anything off the machine that [docs/trust.md](docs/trust.md) says stays
+  on it.
+
+We acknowledge a report within three business days and keep you informed until
+it is resolved. Once a fix is released we credit you in the release notes,
+unless you ask us not to.
+
+## Supported versions
+
+Fixes ship in a new release on PyPI (`stacktrace-cli`). Only the latest release
+is supported: upgrade with `uv tool upgrade stacktrace-cli` or
+`pip install -U stacktrace-cli`. A published version is never re-uploaded, so a
+fix always arrives as a new version number.
+
+## Scope
+
+In scope: the `stacktrace` command, its daemon, and every path by which it
+sends data off the machine (see [docs/trust.md](docs/trust.md)).
+
+Session collection and composition analysis come from two open-source
+dependencies, [`openaidr`](https://github.com/open-agent-security/openaidr) and
+[`openaca`](https://pypi.org/project/openaca/). Report an issue in either one
+to us at the same address; we coordinate the fix upstream.
+
+## Good-faith research
+
+We will not pursue legal action against research that is done in good faith,
+stays within your own machines and accounts, avoids harm to other people's
+data, and gives us reasonable time to fix the issue before disclosure.

{stacktrace_cli-0.11.0 → stacktrace_cli-0.13.0}/docs/adrs/0006-trust-boundary-and-detection-upload.md

@@ -5,7 +5,7 @@ status: accepted
 date: 2026-08-28
 supersedes: null
 superseded-by: null
-amended-by: 0011, 0027, 0031, 0039, 0064, 0084, 0086, 0096, 0109, 0118
+amended-by: 0011, 0027, 0031, 0039, 0064, 0084, 0086, 0096, 0109, 0118, 0138, 0139
 ---
 
 ## Context

{stacktrace_cli-0.11.0 → stacktrace_cli-0.13.0}/docs/adrs/0018-remote-sync-config-and-facade-consumption.md

@@ -6,7 +6,7 @@ date: 2026-09-03
 supersedes: null
 superseded-by: null
 amends: null
-amended-by: 0087
+amended-by: 0087, 0131
 ---
 
 ## Context

{stacktrace_cli-0.11.0 → stacktrace_cli-0.13.0}/docs/adrs/0020-openaca-consumption-boundary.md

@@ -5,6 +5,7 @@ status: accepted
 date: 2026-09-03
 supersedes: null
 superseded-by: null
+amended-by: 0136
 ---
 
 ## Context

{stacktrace_cli-0.11.0 → stacktrace_cli-0.13.0}/docs/adrs/0024-the-upload-carries-observations.md

@@ -6,7 +6,7 @@ date: 2026-09-05
 supersedes: null
 superseded-by: null
 amends: null
-amended-by: null
+amended-by: 0139
 ---
 
 ## Context

{stacktrace_cli-0.11.0 → stacktrace_cli-0.13.0}/docs/adrs/0031-one-host-daemon-observes-native-session-logs.md

@@ -6,7 +6,7 @@ date: 2026-09-17
 supersedes: null
 superseded-by: null
 amends: 0006
-amended-by: 0032, 0036, 0037, 0038, 0039, 0045, 0046, 0066, 0072, 0107, 0113, 0118
+amended-by: 0032, 0036, 0037, 0038, 0039, 0045, 0046, 0066, 0072, 0107, 0113, 0118, 0135
 ---
 
 ## Context

{stacktrace_cli-0.11.0 → stacktrace_cli-0.13.0}/docs/adrs/0070-package-the-cloud-agent-environment-bootstrap.md

@@ -6,7 +6,7 @@ date: 2026-09-25
 supersedes: null
 superseded-by: null
 amends: null
-amended-by: 0123
+amended-by: 0123, 0143
 ---
 
 ## Context

{stacktrace_cli-0.11.0 → stacktrace_cli-0.13.0}/docs/adrs/0072-host-supervisors-own-daemon-restarts.md

@@ -6,7 +6,7 @@ date: 2026-09-25
 supersedes: 0066
 superseded-by: null
 amends: 0031, 0045, 0058
-amended-by: null
+amended-by: 0143
 ---
 
 ## Context

{stacktrace_cli-0.11.0 → stacktrace_cli-0.13.0}/docs/adrs/0073-name-what-fired-and-which-component.md

@@ -6,7 +6,7 @@ date: 2026-09-26
 supersedes: null
 superseded-by: null
 amends: 0067, 0071
-amended-by: 0084, 0085
+amended-by: 0084, 0085, 0138
 ---
 
 ## Context

{stacktrace_cli-0.11.0 → stacktrace_cli-0.13.0}/docs/adrs/0090-count-completed-commands-locate-by-ip-send-through-the-managed-proxy.md

@@ -6,7 +6,7 @@ date: 2026-10-02
 supersedes: null
 superseded-by: null
 amends: 0039
-amended-by: null
+amended-by: 0131
 ---
 
 ## Context

{stacktrace_cli-0.11.0 → stacktrace_cli-0.13.0}/docs/adrs/0093-environment-cost-baselines-count-tokens-keyed-by-model.md

@@ -6,7 +6,7 @@ date: 2026-10-05
 supersedes: null
 superseded-by: null
 amends: null
-amended-by: 0099, 0113, 0114
+amended-by: 0099, 0113, 0114, 0129
 ---
 
 ## Context

{stacktrace_cli-0.11.0 → stacktrace_cli-0.13.0}/docs/adrs/0096-cost-usage-may-leave-as-token-totals-and-work-identifiers.md

@@ -6,7 +6,7 @@ date: 2026-10-06
 supersedes: null
 superseded-by: null
 amends: 0006
-amended-by: 0103, 0109, 0115, 0116
+amended-by: 0103, 0109, 0115, 0116, 0130, 0140
 ---
 
 ## Context

{stacktrace_cli-0.11.0 → stacktrace_cli-0.13.0}/docs/adrs/0099-cost-is-measured-in-tokens-and-relative-increases.md

@@ -6,7 +6,7 @@ date: 2026-10-06
 supersedes: 0094
 superseded-by: null
 amends: 0093
-amended-by: null
+amended-by: 0127
 ---
 
 ## Context

{stacktrace_cli-0.11.0 → stacktrace_cli-0.13.0}/docs/adrs/0101-what-names-a-run-in-fleet.md

@@ -6,7 +6,7 @@ date: 2026-10-08
 supersedes: null
 superseded-by: null
 amends: null
-amended-by: 0109, 0110, 0112, 0121
+amended-by: 0109, 0110, 0112, 0121, 0140
 ---
 
 ## Context

{stacktrace_cli-0.11.0 → stacktrace_cli-0.13.0}/docs/adrs/0104-usage-is-its-own-upload.md

@@ -6,7 +6,7 @@ date: 2026-10-08
 supersedes: null
 superseded-by: null
 amends: 0061
-amended-by: null
+amended-by: 0128
 ---
 
 ## Context

{stacktrace_cli-0.11.0 → stacktrace_cli-0.13.0}/docs/adrs/0109-a-detection-carries-its-runs-title-and-the-persons-name.md

@@ -6,7 +6,7 @@ date: 2026-10-11
 supersedes: null
 superseded-by: null
 amends: 0006, 0096, 0101
-amended-by: 0110, 0112
+amended-by: 0110, 0112, 0140
 ---
 
 ## Context

{stacktrace_cli-0.11.0 → stacktrace_cli-0.13.0}/docs/adrs/0113-size-triggers-behaviour-explains-session-cost-carries-reasons.md

@@ -6,7 +6,7 @@ date: 2026-10-11
 supersedes: null
 superseded-by: null
 amends: 0031, 0092, 0093
-amended-by: null
+amended-by: 0142
 ---
 
 ## Context

{stacktrace_cli-0.11.0 → stacktrace_cli-0.13.0}/docs/adrs/0118-observe-agents-through-adapters-and-a-shared-analyzer.md

@@ -6,7 +6,7 @@ date: 2026-10-11
 supersedes: null
 superseded-by: null
 amends: 0001, 0004, 0006, 0031, 0061
-amended-by: null
+amended-by: 0135
 ---
 
 ## Context

{stacktrace_cli-0.11.0 → stacktrace_cli-0.13.0}/docs/adrs/0123-the-cloud-boot-hook-brings-the-client-up-to-the-latest-release.md

@@ -6,7 +6,7 @@ date: 2026-10-13
 supersedes: null
 superseded-by: null
 amends: 0070
-amended-by: null
+amended-by: 0143
 ---
 
 ## Context

stacktrace_cli-0.13.0/docs/adrs/0127-the-monitor-shows-weighted-tokens-beside-raw-tokens.md

@@ -0,0 +1,121 @@
+---
+id: 0127
+title: The monitor shows weighted tokens beside raw tokens
+status: proposed
+date: 2026-10-15
+supersedes: null
+superseded-by: null
+amends: 0099
+amended-by: null
+---
+
+## Context
+
+The monitor's Usage tab adds fresh input, cache reads, cache writes and output
+one to one. On the measured machine 98% of those tokens were cache reads
+(ADR-0099). A cache read is billed at a tenth of a fresh input token on most
+Claude models, and at less on the newest. So the raw total mostly says how
+often a context was re-read. A long session that re-reads a large cache looks
+heavy. A session that sends the same context uncached on every call looks
+light.
+
+Fleet now judges and shows cost in dollars (Fleet ADR-0087). A developer who
+opens the monitor sees a raw count that moves differently from Fleet's figure.
+Nothing on the page says why.
+
+The product owner asked the monitor to show each kind of token on its own, and
+what each kind weighs, beside the raw count.
+
+ADR-0099 rejected tokens weighted by kind: "that is a price table by another
+name." Its concern was a table kept current for every model, and a figure read
+against an invoice it never matches. That conclusion does not hold for this
+use:
+
+- The table holds ratios between the kinds of one model, as the provider's
+  pricing page states them. It holds no price. The ratios change rarely: every
+  Claude model shares the same cache write and output ratios, and only four
+  models have cheaper cache reads.
+- A model the table does not list still gets a figure, at Anthropic's usual
+  ratios, marked estimated. ADR-0099's "a new model works on day one" holds.
+- A weighted token is still a token: one fresh input token of the model that
+  served it. It names no money and makes no claim about any invoice.
+- Rules, thresholds, baselines, evidence and the usual stay in raw tokens.
+  Only what the monitor shows changes.
+
+## Decision
+
+1. **The monitor shows weighted tokens beside raw tokens.** The Usage tab's
+   head states the range's weighted total under its raw total. A card lists each
+   kind of token with its raw count, the ratio it was weighed at, its weighted
+   count and its share of the weighted total. Each row of the table by model and
+   by agent carries its weighted count. The Live tab's session line carries the
+   session's weighted count beside its tokens.
+2. **One weighted token is one fresh input token of the model that served it.**
+   Each kind is multiplied by that model's published ratio to its own fresh
+   input, model by model, and then summed. On every Anthropic model: fresh input
+   ×1, 5-minute cache write ×1.25, 1-hour cache write ×2, output ×5. Cache reads
+   ×0.1, except ×0.05 on Opus 5.5 and Sonnet 5.5 and ×0.025 on Fable 5.1 and
+   Mythos 5.1. A cache write recorded without its lifetime weighs as a 5-minute
+   write.
+3. **The table lives in the client, as ratios** (`cost/weights.py`). A model is
+   looked up by the name `cost.models` gives it, without its snapshot date, and
+   without the effort suffix Devin records (`claude-opus-5-5-medium`).
+4. **A figure weighed by defaults is marked estimated.** A `claude-*` model the
+   table does not list, any other model (Devin's `swe-1-6`) and a response with
+   no model weigh at Anthropic's usual ratios, cache reads at ×0.1. No token is
+   dropped. The page says "estimated" and names the models.
+5. **Three charges are not weighted:** fast mode, US-only inference, and Haiku
+   5.5's tier for prompts over 100K tokens. Each is billed higher, so each
+   leaves a weighted figure under the bill, never over. The page says so.
+6. **Nothing else changes unit.** The chart, the runs, the components, the
+   usual, Session cost and its reasons, and every figure that leaves the
+   machine stay in raw tokens.
+
+What this amends: ADR-0099's rejection of "tokens weighted by kind", for what
+the monitor shows. ADR-0099's decision stands. Every cost figure is tokens, and
+a weighted token is a token. There is still no price table: this table holds
+ratios.
+
+## Alternatives considered
+
+- **Dollars on the machine.** Rejected by ADR-0099 and by this repo's rule that
+  cost is tokens, never dollars. It needs prices kept current, and invites a
+  comparison with an invoice that subscriptions and discounts never match.
+- **One reference model for every model.** Weigh each model's tokens against a
+  single model's fresh input, so Opus and Haiku add up on one scale. Rejected:
+  the ratio between two models is a ratio of their prices, so this is dollars in
+  disguise, with a reference price to keep current.
+- **Weigh only in Fleet**, as cost-notifications §6 proposed. Rejected: the
+  product owner asked for it on the machine. Fleet now shows dollars (Fleet
+  ADR-0087), so a monitor with raw tokens alone disagrees with Fleet by a factor
+  that moves with the mix of kinds, and nothing explains it.
+- **Show weighted tokens instead of raw ones.** Rejected: the rules, the usual
+  and Session cost read raw tokens, and the page must show the figure they
+  judge.
+- **Fold cache writes without a lifetime into 5-minute writes.** Rejected: that
+  changes the raw 5-minute figure. They get their own row when there are any.
+
+## Consequences
+
+- A reader sees why a large raw total can weigh little: cache reads count a
+  tenth or less.
+- The table is one more thing to keep current. A changed ratio needs a client
+  release. Until it ships, a listed model's figure is off, and a new model's
+  figure says estimated.
+- The machine's weighted tokens and Fleet's dollars are different scales. Within
+  one model they move together. Across models they do not, because each
+  model's weighted token is its own fresh input token.
+- The monitor weighs 1-hour writes at ×2, because transcripts split the two
+  lifetimes. cost-notifications §6 had Fleet weigh both at ×1.25.
+- `tests/cost/test_weights.py` holds each listed model's cache-read ratio,
+  checks that no kind of token is dropped, and checks that every model with a
+  known context window has ratios of its own.
+
+## When to revisit
+
+- When Anthropic changes a ratio, or a new model's cache reads differ from
+  ×0.1.
+- When OpenAIDR reads an agent whose provider publishes different ratios, such
+  as Codex.
+- If Session cost or the usual should judge weighted tokens. That changes the
+  rules, and needs its own ADR.