[model-inventory] Model alias inventory update - 2026-08-01

[github-actions]

Warning

threat detection engine error
The threat detection engine encountered an error and could not complete analysis. This is a tooling failure, not a security finding.

Details

The threat detection engine failed to produce results.

Review the workflow run logs for details.

Summary

Checked live model inventories from OpenAI, Anthropic, Gemini, Copilot SDK, and Copilot (docs
pricing table via Playwright) against pkg/workflow/data/model_aliases.json and the mirrored
models.json pricing payloads (pkg/cli/data/models.json / actions/setup/js/models.json).

• Providers queried: OpenAI, Anthropic, Gemini, Copilot SDK, Copilot (reflect), Copilot docs pricing table
• Total models found: 221 (OpenAI) + 10 (Anthropic) + 50 (Gemini) + 36 (Copilot reflect) + 22 (Copilot SDK) + 35 rows (Copilot docs pricing table, incl. tiered rows)
• Proposed alias changes: 0 (all live github-copilot model IDs are already covered by existing wildcard patterns)
• Pricing gaps found: 3 (corrected in this PR)

No alias updates were needed — this run found and fixed pricing discrepancies in the built-in
models.json payloads instead, which have already been applied directly in the accompanying
commit (not just proposed).

Provider Model Counts

Provider	Models Available	Status
openai	221	✅ ok
anthropic	10	✅ ok
gemini	50	✅ ok
copilot-sdk	22	✅ ok
copilot (reflect)	36	✅ ok
copilot (docs pricing table)	35 rows / 28 unique models	✅ ok (fetched via Playwright, Next.js-rendered)

Raw API Fields Discovered

• OpenAI (raw.json): id, owned_by, created — no pricing/capability fields in the
  /v1/models response itself; pricing must come from external sources.
• Anthropic (raw.json): id, display_name, created_at, type — family grouping via
  display_name (haiku/sonnet/opus) is useful for alias validation; no pricing in-band.
• Gemini (raw.json): input_token_limit, output_token_limit, supported_generation_methods,
  version, description — token limits are a useful proxy for tier inference, already leveraged.
• Copilot SDK (copilot-sdk-models/models.json): rich capabilities.limits (context/output
  windows, vision support with max_prompt_images/supported_media_types), capabilities.supports
  (reasoning effort, thinking budgets, tool calls, streaming), billing.tokenPrices (inputPrice,
  outputPrice, cachePrice, cacheReadPrice, cacheWritePrice, plus longContext tier),
  modelPickerCategory, modelPickerPriceCategory. This is the richest and most authoritative
  pricing source available and was used as primary validation input.
• Copilot reflect (reflect.json): per-model pricing.default / pricing.long_context with
  input/output/cachedInput/cacheWrite/threshold — matches SDK data and was used as a
  cross-check.
• Copilot docs pricing table (Playwright-extracted): Category (Lightweight/Versatile/Powerful),
  Tier (Default/Long context), Threshold (input tokens), Input/Cached input/Output (and
  Cache write for Claude rows) — confirms the SDK/reflect numbers and was the source used to
  identify the 3 discrepancies below.

models.json Pricing Analysis

Missing from models.json

None — all models present in the Copilot SDK, reflect, and docs pricing table are already present
in pkg/cli/data/models.json under github-copilot.

Historical entries not currently returned

gemini-2.5-pro is present in models.json but was not returned by the Copilot SDK, reflect, or
docs pricing table in this run. Retained as a historical entry per policy — not proposed for
removal.

gpt-4.1, gpt-5.2, gpt-5.2-codex are present in models.json but not currently surfaced by
Copilot SDK/reflect/docs (older generations). gpt-4.1 is also one of the intentionally
deprecated/excluded IDs per the task instructions — retained as historical, not removed.

Inferred vs stored pricing discrepancies (fixed in this PR)

Model ID	Stored Pricing (before)	Corrected Pricing	Source
gpt-5.6-luna	input $1.00 / output $6.00 (per M tokens)	input $0.20 / output $1.20	Copilot docs pricing table + reflect pricing.default (both agree)
gpt-5.6-terra	input $2.50 / output $15.00	input $2.00 / output $12.00	Copilot docs pricing table + reflect pricing.default (both agree)
grok-4.5	input $0.20 / output $0.60	input $2.00 / output $6.00	Copilot docs pricing table + reflect pricing.default (both agree)

All other github-copilot model entries (28 checked) matched the docs pricing table and reflect
data exactly — no further discrepancies found. These 3 corrections have already been applied to
both pkg/cli/data/models.json and actions/setup/js/models.json (kept identical) in this
workflow run's commit.

Proposed Alias Updates

None. All live Copilot model IDs from the SDK (claude-sonnet-5, claude-opus-5,
claude-opus-4.8, kimi-k2.7-code, claude-fable-5, gpt-5.6-luna/sol/terra, grok-4.5,
gemini-3.5-flash, gemini-3.6-flash, gemini-3.1-pro-preview, etc.) already resolve correctly
through existing alias patterns (sonnet, opus, fable, kimi, grok, gpt-5.6,
gemini-3.5-flash, gemini-3.6-flash, gemini-3.1-pro). No new model families, no stale
patterns, and no missing task-oriented aliases were identified this run.

Full Model Lists by Provider

Copilot (reflect, 36 total): claude-haiku-4.5, claude-opus-4.6, claude-opus-4.7, claude-opus-4.8,
claude-opus-5, claude-sonnet-4.5, claude-sonnet-4.6, claude-sonnet-5, gpt-3.5-turbo,
gpt-3.5-turbo-0613, gpt-4, gpt-4-0613, gpt-4-o-preview, gpt-4.1, gpt-4.1-2025-04-14,
gpt-41-copilot, gpt-4o, gpt-4o-2024-05-13, gpt-4o-2024-08-06, gpt-4o-2024-11-20, gpt-4o-mini,
gpt-4o-mini-2024-07-18, gpt-5-mini, gpt-5.3-codex, gpt-5.4, gpt-5.4-mini, gpt-5.5, gpt-5.6-luna,
gpt-5.6-sol, gpt-5.6-terra, grok-4.5, mai-code-1-flash-picker, text-embedding-3-small,
text-embedding-3-small-inference, text-embedding-ada-002, trajectory-compaction

Copilot SDK (22 total): auto, claude-haiku-4.5, claude-opus-4.6, claude-opus-4.7,
claude-opus-4.8, claude-opus-5, claude-sonnet-4.5, claude-sonnet-4.6, claude-sonnet-5,
gemini-3.1-pro-preview, gemini-3.5-flash, gemini-3.6-flash, gpt-5-mini, gpt-5.3-codex, gpt-5.4,
gpt-5.4-mini, gpt-5.5, gpt-5.6-luna, gpt-5.6-sol, gpt-5.6-terra, grok-4.5, mai-code-1-flash-picker

models.json github-copilot (32 total): claude-fable-5, claude-haiku-4.5, claude-opus-4.5,
claude-opus-4.6, claude-opus-4.7, claude-opus-4.8, claude-opus-5, claude-sonnet-4,
claude-sonnet-4.5, claude-sonnet-4.6, claude-sonnet-5, gemini-2.5-pro, gemini-3-flash-preview,
gemini-3.1-pro-preview, gemini-3.5-flash, gemini-3.6-flash, gpt-4.1, gpt-5-mini, gpt-5.2,
gpt-5.2-codex, gpt-5.3-codex, gpt-5.4, gpt-5.4-mini, gpt-5.4-nano, gpt-5.5, gpt-5.6-luna,
gpt-5.6-sol, gpt-5.6-terra, grok-4.5, kimi-k2.7-code, mai-code-1-flash-picker, raptor-mini

Notes

• OpenAI and Anthropic direct API keys were not configured for the api-proxy reflect endpoint,
  but the pre-job artifact fetch for these providers succeeded independently (221 and 10 models
  respectively), so their inventories were still available for cross-checking.
• gpt-4o-mini, gpt-4.1, gpt-4o, and gpt-5.4-nano were excluded from the docs pricing table
  per standing policy (intentionally deprecated Copilot-facing IDs) even though gpt-4.1 and
  gpt-4o still appear in the reflect model list.
• models.json already had correct wire_api/provider_type routing metadata for all
  github-copilot entries — no routing metadata changes were needed.

Warning

Firewall blocked 8 domains

The following domains were blocked by the firewall during workflow execution:

• accounts.google.com
• android.clients.google.com
• clients2.google.com
• collector.githubapp.com
• contentautofill.googleapis.com
• proxy.golang.org
• safebrowsingohttpgateway.googleapis.com
• www.google.com

To allow these domains, add them to the network.allowed list in your workflow frontmatter:

network:
  allowed:
    - defaults
    - "accounts.google.com"
    - "android.clients.google.com"
    - "clients2.google.com"
    - "collector.githubapp.com"
    - "contentautofill.googleapis.com"
    - "proxy.golang.org"
    - "safebrowsingohttpgateway.googleapis.com"
    - "www.google.com"

See Network Configuration for more information.

Generated by 📦 Daily Model Inventory Checker · auto · 81.7 AIC · ⊞ 12.5K · ◷

• expires on Aug 7, 2026, 4:20 PM UTC-08:00