coding-agent-cost 0.5.2__tar.gz → 0.5.3__tar.gz

{coding_agent_cost-0.5.2 → coding_agent_cost-0.5.3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: coding-agent-cost
-Version: 0.5.2
+Version: 0.5.3
 Summary: Estimate AI coding agent (Claude Code / Codex CLI) token usage and cost from local logs
 Author: shiki-yusuke
 License: MIT
@@ -17,6 +17,8 @@ Classifier: Programming Language :: Python :: 3.9
 Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Programming Language :: Python :: 3.14
 Classifier: Topic :: Software Development :: Build Tools
 Classifier: Topic :: Utilities
 Requires-Python: >=3.9

{coding_agent_cost-0.5.2 → coding_agent_cost-0.5.3}/agent_cost/__init__.py

@@ -6,4 +6,4 @@ against a versioned rate catalog. Everything happens locally; there are no
 network calls.
 """
 
-__version__ = "0.5.2"
+__version__ = "0.5.3"

{coding_agent_cost-0.5.2 → coding_agent_cost-0.5.3}/agent_cost/rates.json

@@ -1,6 +1,6 @@
 {
   "schema_version": "2",
-  "catalog_version": "2026-10-13",
+  "catalog_version": "2026-11-10",
   "currency": "USD",
   "unit": "per_mtok",
   "usd_per_credit": "0.04",
@@ -14,7 +14,9 @@
     "claude-opus-5-5 added 2026-09-23: rate_id claude-opus-5-5-launch-2026-09-22. effective_from 2026-09-22T00:00:00Z is 00:00:00Z of the official public launch date (anthropic.com/claude-opus-5-5, 'September 22, 2026'); the launch time of day is not published, but a transcript row whose raw model ID is claude-opus-5-5 cannot predate the launch, so pricing from the start of that UTC day cannot price a pre-launch event. Values 4.0/0.20/5.0/8.0/20.0 per the pricing page: cache_read is 0.05x base input (pricing-page footnote 2), a genuine per-model difference from the standard 0.1x and from claude-fable-5-1's 0.025x. fast_multiplier 2.0 (fast mode $8/$40 per MTok on the pricing page). Claude Code's CHANGELOG entry for 2.1.280 states Opus 5.5 became the default model (see sources); rows whose raw model ID is claude-opus-5-5 would be unpriced without this entry. Claude Sonnet 5.5 and Claude Haiku 5.5 are announced to follow 'in the coming weeks' -- add them as separate entries when their model IDs and prices are published, never as aliases of the 5.x entries.",
     "claude-sonnet-5-5 added 2026-09-29: rate_id claude-sonnet-5-5-launch-2026-09-28. effective_from 2026-09-28T00:00:00Z is 00:00:00Z of the official public launch date (anthropic.com/claude-sonnet-5-5, 'September 28, 2026'); as with claude-opus-5-5, the launch time of day is not published, but a transcript row whose raw model ID is claude-sonnet-5-5 cannot predate the launch, so pricing from the start of that UTC day cannot price a pre-launch event. Values 2.0/0.20/2.50/4.0/10.0 per the pricing page (Sonnet 5.5 row, no footnote: cache_read is the standard 0.1x, unlike claude-opus-5-5's 0.05x and claude-fable-5-1's 0.025x). Identical to claude-sonnet-5's ongoing values, but kept as a separate entry, never an alias: the pricing page lists the two rows independently and a future change to one need not apply to the other. fast_multiplier 1.0: the pricing page's fast-mode table lists only the Opus models. Claude Code's CHANGELOG entry for 2.1.284 states Sonnet 5.5 is now the default Sonnet model, so the `sonnet` alias in agent frontmatter resolves to it and rows whose raw model ID is claude-sonnet-5-5 are expected from that version onward; no such row existed in the local transcripts on 2026-09-29. Claude Haiku 5.5 is announced to follow 'in the coming weeks' -- add it as a separate entry when its model ID and prices are published.",
     "claude-haiku-5-5 added 2026-10-09: rate_id claude-haiku-5-5-launch-2026-10-07. effective_from 2026-10-07T00:00:00Z is 00:00:00Z of the official release date (platform.claude.com/docs/en/models/haiku-5-5/overview, 'Released October 7, 2026'); as with claude-sonnet-5-5, the release time of day is not published, but a transcript row whose raw model ID is claude-haiku-5-5 cannot predate the release. Its prices are tiered by prompt length: the period's top-level values (0.10/0.01/0.125/0.20/0.50) apply to prompts up to 100,000 tokens, and prompt_tiers[0] (prompt_tokens_over 100000: 0.50/0.05/0.625/1.0/2.50) to prompts over 100,000 tokens, per the pricing page's two Haiku 5.5 rows (no footnote: cache_read is the standard 0.1x). Prompt length follows the pricing page's Long context pricing section: it counts all of a request's input tokens including cache reads and cache writes, and each request is priced on its own, so a request over the threshold pays the higher prices for every token kind, output included. A fact whose prompt length is unknown (Codex, a conflicting Claude dedup group, inconsistent usage) is priced at the top-level values as a lower_bound. A separate entry, never an alias of claude-haiku-4-5. fast_multiplier 1.0: the pricing page's fast-mode table lists only the Opus models. Claude Code's CHANGELOG entry for 2.1.293 states Haiku 5.5 is now the default Haiku model, so the `haiku` alias resolves to it from that version onward. schema_version is 2 because prompt_tiers is new: agent-cost 0.3.0 and earlier reject schema_version 2 as unsupported instead of silently pricing every claude-haiku-5-5 request at the cheaper tier, and a schema_version 1 catalog may not carry prompt_tiers.",
-    "claude-sonnet-5-5 cache_read cut 2026-10-13: the launch period (rate_id claude-sonnet-5-5-launch-2026-09-28) now ends at 2026-10-07T00:00:00Z and a new period (rate_id claude-sonnet-5-5-cache-read-cut-2026-10-07) prices cache_read at 0.10 instead of 0.20 from that instant; input_nocache, cache_write_5m, cache_write_1h and output are unchanged (2.0/2.50/4.0/10.0). anthropic.com/claude-haiku-5-5 (dated October 7, 2026) states 'starting today, we're lowering the price of cache reads on Claude Sonnet 5.5' and 'Cache reads now cost 50% less: $0.10 per million tokens rather than $0.20'; the source gives a calendar date only, so effective_from is 00:00:00Z of that day, following the claude-sonnet-5-standard-2026-09-01 precedent. The pricing page (retrieved 2026-10-13) lists the Sonnet 5.5 row as $2 / $2.50 / $4 / $0.10 / $10 with footnote 2: cache hits and refreshes on Claude Opus 5.5 and Claude Sonnet 5.5 are 0.05x the base input price, so cache_read is now 0.05x like claude-opus-5-5. Claude Code's CHANGELOG entry for 2.1.296 (2026-10-09) updated /cost, the status line, --max-budget-usd and the SDK's cost figures to $0.10 per million Sonnet 5.5 cache-read tokens; that is the date Claude Code's own figures changed, not the API price's effective date. catalog 2026-10-09 priced claude-sonnet-5-5 cache_read at 0.20 for events from 2026-10-07 through 2026-10-13, overstating that component 2x; reports produced with it for windows touching 2026-10-07 or later should be regenerated. claude-sonnet-5 is unchanged (cache_read stays 0.20): the pricing page's footnote 3 covers its input/output prices only, and footnote 2 names Opus 5.5 and Sonnet 5.5, not Sonnet 5."
+    "claude-sonnet-5-5 cache_read cut 2026-10-13: the launch period (rate_id claude-sonnet-5-5-launch-2026-09-28) now ends at 2026-10-07T00:00:00Z and a new period (rate_id claude-sonnet-5-5-cache-read-cut-2026-10-07) prices cache_read at 0.10 instead of 0.20 from that instant; input_nocache, cache_write_5m, cache_write_1h and output are unchanged (2.0/2.50/4.0/10.0). anthropic.com/claude-haiku-5-5 (dated October 7, 2026) states 'starting today, we're lowering the price of cache reads on Claude Sonnet 5.5' and 'Cache reads now cost 50% less: $0.10 per million tokens rather than $0.20'; the source gives a calendar date only, so effective_from is 00:00:00Z of that day, following the claude-sonnet-5-standard-2026-09-01 precedent. The pricing page (retrieved 2026-10-13) lists the Sonnet 5.5 row as $2 / $2.50 / $4 / $0.10 / $10 with footnote 2: cache hits and refreshes on Claude Opus 5.5 and Claude Sonnet 5.5 are 0.05x the base input price, so cache_read is now 0.05x like claude-opus-5-5. Claude Code's CHANGELOG entry for 2.1.296 (2026-10-09) updated /cost, the status line, --max-budget-usd and the SDK's cost figures to $0.10 per million Sonnet 5.5 cache-read tokens; that is the date Claude Code's own figures changed, not the API price's effective date. catalog 2026-10-09 priced claude-sonnet-5-5 cache_read at 0.20 for events from 2026-10-07 through 2026-10-13, overstating that component 2x; reports produced with it for windows touching 2026-10-07 or later should be regenerated. claude-sonnet-5 is unchanged (cache_read stays 0.20): the pricing page's footnote 3 covers its input/output prices only, and footnote 2 names Opus 5.5 and Sonnet 5.5, not Sonnet 5.",
+    "gpt-6.1-terra, gpt-6.1-sol and codex-auto-review added 2026-11-09: standard tier, short context USD per MTok from the human-approved price table pricing-2026-11-v4 (developers.openai.com/api/docs/pricing, retrieved 2026-11-05); effective_from 2026-10-01T00:00:00Z is a placeholder before the first observed use (official launch date unconfirmed; rate_id *-launch-date-unconfirmed, following the claude-opus-5 placeholder precedent); usage from 2026-11-04 must be priced, so the retrieval date 2026-11-05 cannot be used. codex-auto-review is the model name of the Codex automatic review (guardian subagent) and has no published price; a human decision on 2026-11-06 prices it the same as gpt-6.1-terra. credits_per_mtok follow usd_per_credit=0.04. Existing entries (including gpt-5.6-terra) are unchanged.",
+    "claude-fable-5-2 and claude-mythos-5-2 added 2026-11-10: rate_ids claude-fable-5-2-launch-2026-11-09 / claude-mythos-5-2-launch-2026-11-09. effective_from 2026-11-09T00:00:00Z is 00:00:00Z of the official announcement date (anthropic.com/claude-fable-and-mythos-5-2, published November 9, 2026); the launch time of day is not published, but a transcript row whose raw model ID is claude-fable-5-2 or claude-mythos-5-2 cannot predate the launch. Values 8.0/0.20/10.0/16.0/40.0 per the pricing page (retrieved 2026-11-10): cache_read is 0.025x base input (footnote 1), like claude-fable-5-1. claude-mythos-5-2 (limited availability, verified organizations only) has the same prices; it is kept as a separate entry, never an alias of claude-fable-5-2. fast_multiplier 2.0 (fast mode $16/$80 per MTok on the pricing page). No prompt-length tiers (1M context at standard prices). Claude Code's CHANGELOG entry for 2.1.321 adds claude-fable-5-2 at $8/$40 with $0.20 cache reads. claude-fable-5-1 is unchanged (10.0/0.25/12.50/20.0/50.0). claude-mythos-5-1 stays absent (official release date unconfirmed), so its rows remain unpriced."
   ],
   "sources": [
     {
@@ -124,6 +126,20 @@
       "url": "https://www.anthropic.com/claude-haiku-5-5",
       "retrieved_at": "2026-10-13",
       "note": "Haiku 5.5 announcement dated October 7, 2026, 'Further updates': starting that day, Claude Sonnet 5.5 cache reads cost $0.10 per million tokens rather than $0.20; the effective date of the claude-sonnet-5-5 cache_read cut"
+    },
+    {
+      "url": "https://developers.openai.com/api/docs/pricing",
+      "retrieved_at": "2026-11-05"
+    },
+    {
+      "url": "https://www.anthropic.com/claude-fable-and-mythos-5-2",
+      "retrieved_at": "2026-11-10",
+      "note": "Fable 5.2 / Mythos 5.2 announcement published November 9, 2026; the effective date of claude-fable-5-2 and claude-mythos-5-2"
+    },
+    {
+      "url": "https://platform.claude.com/docs/en/about-claude/pricing",
+      "retrieved_at": "2026-11-10",
+      "note": "Fable 5.2 / Mythos 5.2 rows: $8 input, $10 5m cache write, $16 1h cache write, $0.20 cache hits (footnote 1: 0.025x), $40 output; fast mode $16/$80"
     }
   ],
   "models": [
@@ -161,6 +177,23 @@
         }
       ]
     },
+    {
+      "model_key": "claude-fable-5-2",
+      "aliases": [],
+      "fast_multiplier": "2.0",
+      "rates": [
+        {
+          "rate_id": "claude-fable-5-2-launch-2026-11-09",
+          "effective_from": "2026-11-09T00:00:00+00:00",
+          "effective_until": null,
+          "input_nocache": "8.0",
+          "cache_read": "0.20",
+          "cache_write_5m": "10.0",
+          "cache_write_1h": "16.0",
+          "output": "40.0"
+        }
+      ]
+    },
     {
       "model_key": "claude-mythos-5",
       "aliases": [],
@@ -178,6 +211,23 @@
         }
       ]
     },
+    {
+      "model_key": "claude-mythos-5-2",
+      "aliases": [],
+      "fast_multiplier": "2.0",
+      "rates": [
+        {
+          "rate_id": "claude-mythos-5-2-launch-2026-11-09",
+          "effective_from": "2026-11-09T00:00:00+00:00",
+          "effective_until": null,
+          "input_nocache": "8.0",
+          "cache_read": "0.20",
+          "cache_write_5m": "10.0",
+          "cache_write_1h": "16.0",
+          "output": "40.0"
+        }
+      ]
+    },
     {
       "model_key": "claude-opus-4-8",
       "aliases": [],
@@ -689,6 +739,78 @@
           "output": "50.0"
         }
       ]
+    },
+    {
+      "model_key": "gpt-6.1-terra",
+      "aliases": [],
+      "fast_multiplier": "1.0",
+      "credits_per_mtok": {
+        "input_nocache": "25.0",
+        "cache_read": "1.25",
+        "cache_write_5m": null,
+        "cache_write_1h": null,
+        "output": "125.0"
+      },
+      "rates": [
+        {
+          "rate_id": "gpt-6.1-terra-launch-date-unconfirmed",
+          "effective_from": "2026-10-01T00:00:00+00:00",
+          "effective_until": null,
+          "input_nocache": "1.00",
+          "cache_read": "0.05",
+          "cache_write_5m": null,
+          "cache_write_1h": null,
+          "output": "5.00"
+        }
+      ]
+    },
+    {
+      "model_key": "gpt-6.1-sol",
+      "aliases": [],
+      "fast_multiplier": "1.0",
+      "credits_per_mtok": {
+        "input_nocache": "50.0",
+        "cache_read": "2.5",
+        "cache_write_5m": null,
+        "cache_write_1h": null,
+        "output": "250.0"
+      },
+      "rates": [
+        {
+          "rate_id": "gpt-6.1-sol-launch-date-unconfirmed",
+          "effective_from": "2026-10-01T00:00:00+00:00",
+          "effective_until": null,
+          "input_nocache": "2.00",
+          "cache_read": "0.10",
+          "cache_write_5m": null,
+          "cache_write_1h": null,
+          "output": "10.00"
+        }
+      ]
+    },
+    {
+      "model_key": "codex-auto-review",
+      "aliases": [],
+      "fast_multiplier": "1.0",
+      "credits_per_mtok": {
+        "input_nocache": "25.0",
+        "cache_read": "1.25",
+        "cache_write_5m": null,
+        "cache_write_1h": null,
+        "output": "125.0"
+      },
+      "rates": [
+        {
+          "rate_id": "codex-auto-review-2026-11-terra-equivalent",
+          "effective_from": "2026-10-01T00:00:00+00:00",
+          "effective_until": null,
+          "input_nocache": "1.00",
+          "cache_read": "0.05",
+          "cache_write_5m": null,
+          "cache_write_1h": null,
+          "output": "5.00"
+        }
+      ]
     }
   ]
 }

{coding_agent_cost-0.5.2 → coding_agent_cost-0.5.3}/coding_agent_cost.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: coding-agent-cost
-Version: 0.5.2
+Version: 0.5.3
 Summary: Estimate AI coding agent (Claude Code / Codex CLI) token usage and cost from local logs
 Author: shiki-yusuke
 License: MIT
@@ -17,6 +17,8 @@ Classifier: Programming Language :: Python :: 3.9
 Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Programming Language :: Python :: 3.14
 Classifier: Topic :: Software Development :: Build Tools
 Classifier: Topic :: Utilities
 Requires-Python: >=3.9

{coding_agent_cost-0.5.2 → coding_agent_cost-0.5.3}/coding_agent_cost.egg-info/SOURCES.txt

@@ -24,7 +24,9 @@ tests/test_billing_plan.py
 tests/test_cli.py
 tests/test_e0a_fable_5_1_and_sonnet_5_correction.py
 tests/test_e0a_review3_hardening.py
+tests/test_fable_5_2_rates.py
 tests/test_facts.py
+tests/test_gpt_6_1_rates.py
 tests/test_haiku_5_5_rates.py
 tests/test_measure_v1_contract.py
 tests/test_opus_5_5_rates.py

{coding_agent_cost-0.5.2 → coding_agent_cost-0.5.3}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "coding-agent-cost"
-version = "0.5.2"
+version = "0.5.3"
 description = "Estimate AI coding agent (Claude Code / Codex CLI) token usage and cost from local logs"
 readme = "README.md"
 license = { text = "MIT" }
@@ -31,6 +31,8 @@ classifiers = [
     "Programming Language :: Python :: 3.10",
     "Programming Language :: Python :: 3.11",
     "Programming Language :: Python :: 3.12",
+    "Programming Language :: Python :: 3.13",
+    "Programming Language :: Python :: 3.14",
     "Topic :: Software Development :: Build Tools",
     "Topic :: Utilities",
 ]

{coding_agent_cost-0.5.2 → coding_agent_cost-0.5.3}/tests/test_astra_pricing.py

@@ -45,7 +45,7 @@ def test_standard_fast_and_unknown_mode_prices(kind, usd, credits, mode, multipl
 
 def test_catalog_observation_boundary_is_not_backdated():
     catalog = load_rates()
-    assert catalog.catalog_version == "2026-10-13"
+    assert catalog.catalog_version == "2026-11-10"
     entry = catalog.models["gpt-6-astra"]
     assert entry.aliases == ()
     assert entry.rates[0].effective_from == CUTOFF
@@ -149,7 +149,7 @@ def test_measure_reads_isolated_synthetic_astra_with_existing_contract(tmp_path,
     assert cli.main(["measure", "--session-id", "synthetic-astra", "--agent", "codex"]) == 0
     report = json.loads(capsys.readouterr().out)
     assert report["protocol_version"] == "measure/v1"
-    assert report["rates"]["catalog_version"] == "2026-10-13"
+    assert report["rates"]["catalog_version"] == "2026-11-10"
     assert len(report["rates"]["sha256"]) == 64
     assert report["sessions"]["synthetic-astra"]["totals"]["estimated_cost_usd"] == 122
     assert report["total"]["totals"]["credits"] == 3050

{coding_agent_cost-0.5.2 → coding_agent_cost-0.5.3}/tests/test_billing_plan.py

@@ -13,6 +13,8 @@ import copy
 import hashlib
 import json
 import os
+
+_REAL_FSTAT = os.fstat
 import traceback
 from datetime import datetime, timezone
 from decimal import Decimal
@@ -583,11 +585,24 @@ def test_realpath_oserror_is_billing_plan_error(tmp_path, monkeypatch):
     _assert_no_leak(info.value, path)
 
 
+def _is_fd_of(fd, path):
+    """True when fd refers to path. Keeps fakes scoped to the plan file: pytest's
+    capture machinery calls os.fstat/os.read/os.close on its own fds (pytest 9.2+)."""
+    try:
+        st = _REAL_FSTAT(fd)
+    except OSError:
+        return False
+    target = os.stat(path)
+    return (st.st_dev, st.st_ino) == (target.st_dev, target.st_ino)
+
+
 def test_fstat_oserror_is_billing_plan_error(tmp_path, monkeypatch):
     path = _write_plan(tmp_path, _base_plan())
 
     def failing_fstat(fd):
-        raise _oserror_on(path)
+        if _is_fd_of(fd, path):
+            raise _oserror_on(path)
+        return _REAL_FSTAT(fd)
 
     monkeypatch.setattr(bp.os, "fstat", failing_fstat)
     with pytest.raises(bp.BillingPlanError) as info:
@@ -598,8 +613,12 @@ def test_fstat_oserror_is_billing_plan_error(tmp_path, monkeypatch):
 def test_read_oserror_is_billing_plan_error(tmp_path, monkeypatch):
     path = _write_plan(tmp_path, _base_plan())
 
+    real_read = os.read
+
     def failing_read(fd, n):
-        raise _oserror_on(path)
+        if _is_fd_of(fd, path):
+            raise _oserror_on(path)
+        return real_read(fd, n)
 
     monkeypatch.setattr(bp.os, "read", failing_read)
     with pytest.raises(bp.BillingPlanError) as info:
@@ -612,8 +631,10 @@ def test_close_oserror_is_billing_plan_error(tmp_path, monkeypatch):
     real_close = os.close
 
     def failing_close(fd):
+        mine = _is_fd_of(fd, path)
         real_close(fd)
-        raise _oserror_on(path)
+        if mine:
+            raise _oserror_on(path)
 
     monkeypatch.setattr(bp.os, "close", failing_close)
     with pytest.raises(bp.BillingPlanError) as info:
@@ -626,8 +647,10 @@ def test_close_oserror_does_not_replace_an_earlier_error(tmp_path, monkeypatch):
     real_close = os.close
 
     def failing_close(fd):
+        mine = _is_fd_of(fd, path)
         real_close(fd)
-        raise _oserror_on(path)
+        if mine:
+            raise _oserror_on(path)
 
     monkeypatch.setattr(bp.os, "close", failing_close)
     with pytest.raises(bp.BillingPlanError) as info:

{coding_agent_cost-0.5.2 → coding_agent_cost-0.5.3}/tests/test_cli.py

@@ -886,7 +886,7 @@ def test_rates_show_model_without_tiers_prints_no_prompt_tiers(capsys):
 def test_rates_validate_packaged_catalog_with_prompt_tiers_exits_zero(capsys):
     rc = cli.main(["rates", "validate"])
     assert rc == 0
-    assert "catalog_version=2026-10-13" in capsys.readouterr().out
+    assert "catalog_version=2026-11-10" in capsys.readouterr().out
 
 
 # ---------------------------------------------------------------------------

{coding_agent_cost-0.5.2 → coding_agent_cost-0.5.3}/tests/test_e0a_review3_hardening.py

@@ -19,7 +19,7 @@ architect レビュー (3 巡目) と builder レビューで、既存テスト
        かつ確定金額が計上されない。境界ちょうどは priced になる。
 
 条件D (PR レビュー3点目・追加分): catalog は
-       catalog_version == "2026-10-13"（rates.json の現行値）であり、sources の中に url
+       catalog_version == "2026-11-10"（rates.json の現行値）であり、sources の中に url
        "https://platform.claude.com/docs/en/about-claude/pricing" の
        entry があって、その note に
        "sha256=d79ad28567196bd55dd50e2fd89341b9da9774a45c1d02fcf387078507fd15e0"
@@ -224,7 +224,7 @@ def test_condition_c_claude_fable_5_1_at_effective_from_boundary_is_priced_contr
 
 
 def test_condition_d_catalog_version_and_pricing_source_sha256_note():
-    """条件D: catalog_version == "2026-10-13"（rates.json の現行値）であり、sources に url
+    """条件D: catalog_version == "2026-11-10"（rates.json の現行値）であり、sources に url
     "https://platform.claude.com/docs/en/about-claude/pricing" の entry が
     あって、その note に指定の sha256 文字列を含むこと。
 
@@ -233,7 +233,7 @@ def test_condition_d_catalog_version_and_pricing_source_sha256_note():
     ダイジェストを貼り付けてしまう。
     """
     catalog = load_rates()
-    assert catalog.catalog_version == "2026-10-13"
+    assert catalog.catalog_version == "2026-11-10"
 
     matches = [
         s

coding_agent_cost-0.5.3/tests/test_fable_5_2_rates.py

@@ -0,0 +1,83 @@
+"""Acceptance tests for claude-fable-5-2 / claude-mythos-5-2 (catalog_version 2026-11-10).
+
+Source: platform.claude.com/docs/en/about-claude/pricing (retrieved 2026-11-10) and the
+announcement anthropic.com/claude-fable-and-mythos-5-2 (published 2026-11-09).
+"""
+
+from datetime import datetime
+from decimal import Decimal
+
+from agent_cost.rates import load_rates
+
+USD = {
+    "input_nocache": "8.0",
+    "cache_read": "0.20",
+    "cache_write_5m": "10.0",
+    "cache_write_1h": "16.0",
+    "output": "40.0",
+}
+MODELS = ("claude-fable-5-2", "claude-mythos-5-2")
+
+
+def dt(s):
+    return datetime.fromisoformat(s)
+
+
+def test_fable_and_mythos_5_2_resolve_with_pricing_page_values_from_launch():
+    catalog = load_rates()
+    for model in MODELS:
+        for occurred_at in ("2026-11-09T00:00:00+00:00", "2026-11-10T03:00:00+00:00", "2027-06-01T00:00:00+00:00"):
+            resolved, period = catalog.rate_for(model, dt(occurred_at))
+            assert resolved == model
+            assert period is not None, (model, occurred_at)
+            assert period.rate_id == f"{model}-launch-2026-11-09"
+            assert period.effective_until is None
+            assert period.prompt_tiers == ()
+            for kind, value in USD.items():
+                assert period.values[kind] == Decimal(value), (model, kind)
+
+
+def test_fable_and_mythos_5_2_are_unpriced_before_launch():
+    catalog = load_rates()
+    for model in MODELS:
+        resolved, period = catalog.rate_for(model, dt("2026-11-08T23:59:59+00:00"))
+        assert resolved == model
+        assert period is None
+
+
+def test_cache_read_is_0_025x_and_fast_mode_is_2x():
+    catalog = load_rates()
+    for model in MODELS:
+        entry = catalog.models[model]
+        assert entry.aliases == ()
+        assert entry.fast_multiplier == Decimal("2.0")  # fast mode $16 / $80
+        assert len(entry.rates) == 1
+        _, period = catalog.rate_for(model, dt("2026-11-10T00:00:00+00:00"))
+        assert period.values["cache_read"] == period.values["input_nocache"] * Decimal("0.025")
+
+
+def test_separate_entries_not_aliases():
+    catalog = load_rates()
+    for model in MODELS:
+        for other in catalog.models.values():
+            assert model not in other.aliases
+
+
+def test_fable_5_1_is_unchanged_and_mythos_5_1_stays_unpriced():
+    catalog = load_rates()
+    _, period = catalog.rate_for("claude-fable-5-1", dt("2026-11-10T00:00:00+00:00"))
+    assert period.rate_id == "claude-fable-5-1-local-first-observed-2026-09-02"
+    assert period.values["input_nocache"] == Decimal("10.0")
+    assert period.values["cache_read"] == Decimal("0.25")
+    assert period.values["output"] == Decimal("50.0")
+    assert "claude-mythos-5-1" not in catalog.models
+    _, period = catalog.rate_for("claude-mythos-5-1", dt("2026-11-10T00:00:00+00:00"))
+    assert period is None
+
+
+def test_catalog_version_and_sources_record_fable_5_2():
+    catalog = load_rates()
+    assert catalog.catalog_version == "2026-11-10"
+    urls = [(s.get("url"), s.get("retrieved_at")) for s in catalog.sources]
+    assert ("https://www.anthropic.com/claude-fable-and-mythos-5-2", "2026-11-10") in urls
+    assert ("https://platform.claude.com/docs/en/about-claude/pricing", "2026-11-10") in urls

coding_agent_cost-0.5.3/tests/test_gpt_6_1_rates.py

@@ -0,0 +1,117 @@
+"""Acceptance tests for gpt-6.1-terra / gpt-6.1-sol / codex-auto-review (catalog_version 2026-11-09).
+
+Source: human-approved price table pricing-2026-11-v4 (developers.openai.com/api/docs/pricing,
+retrieved 2026-11-05; standard tier, short context). codex-auto-review (Codex automatic
+review / guardian subagent) has no published price; priced as gpt-6.1-terra by a human
+decision on 2026-11-06.
+"""
+
+import json
+from datetime import datetime, timezone
+from decimal import Decimal
+from pathlib import Path
+
+import agent_cost
+from agent_cost.rates import load_rates
+
+UTC = timezone.utc
+
+EXPECTED = {
+    "gpt-6.1-terra": (
+        "gpt-6.1-terra-launch-date-unconfirmed",
+        {"input_nocache": "1.00", "cache_read": "0.05", "output": "5.00"},
+        {"input_nocache": "25.0", "cache_read": "1.25", "output": "125.0"},
+    ),
+    "gpt-6.1-sol": (
+        "gpt-6.1-sol-launch-date-unconfirmed",
+        {"input_nocache": "2.00", "cache_read": "0.10", "output": "10.00"},
+        {"input_nocache": "50.0", "cache_read": "2.5", "output": "250.0"},
+    ),
+    "codex-auto-review": (
+        "codex-auto-review-2026-11-terra-equivalent",
+        {"input_nocache": "1.00", "cache_read": "0.05", "output": "5.00"},
+        {"input_nocache": "25.0", "cache_read": "1.25", "output": "125.0"},
+    ),
+}
+
+
+def dt(s):
+    return datetime.fromisoformat(s)
+
+
+def test_gpt_6_1_models_resolve_with_pricing_page_values_from_effective_from():
+    catalog = load_rates()
+    for model, (rate_id, usd, _credits) in EXPECTED.items():
+        for occurred_at in ("2026-10-01T00:00:00+00:00", "2026-11-04T07:49:42+00:00", "2027-01-01T00:00:00+00:00"):
+            resolved, period = catalog.rate_for(model, dt(occurred_at))
+            assert resolved == model
+            assert period is not None, (model, occurred_at)
+            assert period.rate_id == rate_id
+            assert period.effective_until is None
+            for kind, value in usd.items():
+                assert period.values[kind] == Decimal(value), (model, kind)
+            assert period.values["cache_write_5m"] is None
+            assert period.values["cache_write_1h"] is None
+            assert period.prompt_tiers == ()
+
+
+def test_gpt_6_1_models_are_unpriced_before_effective_from():
+    catalog = load_rates()
+    for model in EXPECTED:
+        resolved, period = catalog.rate_for(model, dt("2026-09-30T23:59:59+00:00"))
+        assert resolved == model
+        assert period is None
+
+
+def test_gpt_6_1_models_credits_follow_usd_per_credit_and_entry_shape():
+    catalog = load_rates()
+    usd_per_credit = Decimal(catalog.usd_per_credit)
+    for model, (_rate_id, usd, credits) in EXPECTED.items():
+        entry = catalog.models[model]
+        assert entry.aliases == ()
+        assert entry.fast_multiplier == Decimal("1.0")
+        assert entry.credits_per_mtok["cache_write_5m"] is None
+        assert entry.credits_per_mtok["cache_write_1h"] is None
+        for kind, value in credits.items():
+            assert entry.credits_per_mtok[kind] == Decimal(value), (model, kind)
+            assert entry.credits_per_mtok[kind] * usd_per_credit == Decimal(usd[kind])
+
+
+def test_codex_auto_review_is_priced_like_gpt_6_1_terra_but_is_its_own_entry():
+    catalog = load_rates()
+    at = dt("2026-11-06T00:00:00+00:00")
+    _, terra = catalog.rate_for("gpt-6.1-terra", at)
+    _, review = catalog.rate_for("codex-auto-review", at)
+    assert review.values == terra.values
+    assert review.rate_id != terra.rate_id
+    assert "codex-auto-review" not in catalog.models["gpt-6.1-terra"].aliases
+
+
+def test_unknown_gpt_6_1_luna_stays_unpriced():
+    catalog = load_rates()
+    assert "gpt-6.1-luna" not in catalog.models
+    resolved, period = catalog.rate_for("gpt-6.1-luna", dt("2026-11-06T00:00:00+00:00"))
+    assert period is None
+
+
+def test_existing_gpt_5_6_terra_values_are_unchanged():
+    catalog = load_rates()
+    entry = catalog.models["gpt-5.6-terra"]
+    assert len(entry.rates) == 1
+    _, period = catalog.rate_for("gpt-5.6-terra", dt("2026-11-06T00:00:00+00:00"))
+    assert period.rate_id == "gpt-5.6-terra-2026-07-secondary-sourced"
+    assert period.values["input_nocache"] == Decimal("2.50")
+    assert period.values["cache_read"] == Decimal("0.25")
+    assert period.values["output"] == Decimal("15.0")
+    assert entry.credits_per_mtok["input_nocache"] == Decimal("62.5")
+    assert entry.credits_per_mtok["cache_read"] == Decimal("6.25")
+    assert entry.credits_per_mtok["output"] == Decimal("375.0")
+
+
+def test_catalog_version_and_source_record_the_gpt_6_1_pricing_page():
+    catalog = load_rates()
+    assert catalog.catalog_version >= "2026-11-09"
+    urls = [(s.get("url"), s.get("retrieved_at")) for s in catalog.sources]
+    assert urls.count(("https://developers.openai.com/api/docs/pricing", "2026-11-05")) == 1
+    raw = json.loads((Path(agent_cost.__file__).parent / "rates.json").read_text(encoding="utf-8"))
+    assert any("codex-auto-review" in n and "2026-11-06" in n for n in raw["notes"])

{coding_agent_cost-0.5.2 → coding_agent_cost-0.5.3}/tests/test_haiku_5_5_rates.py

@@ -79,7 +79,7 @@ def test_claude_haiku_5_5_cache_multipliers_are_standard_in_every_tier_and_no_fa
 
 def test_catalog_version_schema_version_and_sources_record_the_haiku_5_5_primary_sources():
     catalog = load_rates()
-    assert catalog.catalog_version == "2026-10-13"
+    assert catalog.catalog_version == "2026-11-10"
     assert catalog.schema_version == "2"
     urls = {(s.get("url"), s.get("retrieved_at")) for s in catalog.sources}
     assert ("https://platform.claude.com/docs/en/about-claude/pricing", "2026-10-09") in urls

{coding_agent_cost-0.5.2 → coding_agent_cost-0.5.3}/tests/test_rates.py

@@ -527,6 +527,7 @@ def test_packaged_catalog_existing_models_are_unchanged():
     rows are never changed in place; a price change adds a new period.
     """
     raw = json.loads(resources.files("agent_cost").joinpath("rates.json").read_bytes())
-    existing = [m for m in raw["models"] if m["model_key"] != "claude-haiku-5-5"]
+    added_after = {"claude-haiku-5-5", "gpt-6.1-terra", "gpt-6.1-sol", "codex-auto-review", "claude-fable-5-2", "claude-mythos-5-2"}
+    existing = [m for m in raw["models"] if m["model_key"] not in added_after]
     digest = hashlib.sha256(json.dumps(existing, sort_keys=True, separators=(",", ":")).encode()).hexdigest()
     assert digest == _EXISTING_MODELS_SHA256

{coding_agent_cost-0.5.2 → coding_agent_cost-0.5.3}/tests/test_sonnet_5_5_rates.py

@@ -112,7 +112,7 @@ def test_claude_sonnet_5_5_cache_read_multiplier_per_period_and_no_fast_mode():
 
 def test_catalog_version_and_sources_record_the_sonnet_5_5_primary_sources():
     catalog = load_rates()
-    assert catalog.catalog_version == "2026-10-13"
+    assert catalog.catalog_version == "2026-11-10"
     urls = {(s.get("url"), s.get("retrieved_at")) for s in catalog.sources}
     assert ("https://platform.claude.com/docs/en/about-claude/pricing", "2026-09-29") in urls
     assert ("https://www.anthropic.com/claude-sonnet-5-5", "2026-09-29") in urls

{coding_agent_cost-0.5.2 → coding_agent_cost-0.5.3}/tests/test_version_consistency.py

@@ -1,8 +1,8 @@
 """Spec: pyproject.toml's [project].version and agent_cost.__version__ must
-agree, and both must read "0.5.2" for this change (the packaged catalog
-lowers claude-sonnet-5-5 cache_read to $0.10 from 2026-10-07 -- a price
-correction with no code or schema change, so it bumps the patch version). A
-broken
+agree, and both must read "0.5.3" for this change (the packaged catalog adds
+claude-fable-5-2, claude-mythos-5-2, gpt-6.1-terra, gpt-6.1-sol and
+codex-auto-review -- catalog entries only, with no code or schema change, so
+it bumps the patch version). A broken
 implementation would bump one file but not the other, or forget the bump
 entirely.
 
@@ -31,5 +31,5 @@ def test_pyproject_version_matches_package_version():
 
 
 def test_version_is_0_5_2():
-    assert agent_cost.__version__ == "0.5.2"
-    assert _pyproject_version() == "0.5.2"
+    assert agent_cost.__version__ == "0.5.3"
+    assert _pyproject_version() == "0.5.3"