fix: resolve three open bugs (#2836, #2858, #2873)

- fix(context-limit): check modelContextLimitsCache for all Anthropic
  models, not just GA-model set; user config/cache wins over 200K default
  (fixes #2836)

- fix(agent-key-remapper): preserve config key aliases alongside display
  names so `opencode run --agent sisyphus` resolves correctly
  (fixes #2858)

- fix(tool-config): respect host permission.skill=deny by disabling
  skill/skill_mcp tools when host denies them (fixes #2873)

- test: update context-limit and agent-key-remapper tests to match new
  behavior

src/shared/context-limit-resolver.test.ts

@@ -45,7 +45,7 @@ describe("resolveActualContextLimit", () => {
     expect(actualLimit).toBe(1_000_000)
   })
 
-  it("returns default 200K for older Anthropic models even when cached limit is higher", () => {
+  it("returns cached limit for Anthropic models when modelContextLimitsCache has entry", () => {
     // given
     delete process.env[ANTHROPIC_CONTEXT_ENV_KEY]
     delete process.env[VERTEX_CONTEXT_ENV_KEY]
@@ -59,7 +59,7 @@ describe("resolveActualContextLimit", () => {
     })
 
     // then
-    expect(actualLimit).toBe(200_000)
+    expect(actualLimit).toBe(500_000)
   })
 
   it("returns default 200K for Anthropic models without cached limit and 1M mode disabled", () => {

src/shared/context-limit-resolver.ts

@@ -1,13 +1,6 @@
 import process from "node:process"
 
 const DEFAULT_ANTHROPIC_ACTUAL_LIMIT = 200_000
-const ANTHROPIC_NO_HEADER_GA_MODEL_IDS = new Set([
-  "claude-opus-4-6",
-  "claude-opus-4.6",
-  "claude-sonnet-4-6",
-  "claude-sonnet-4.6",
-])
-
 export type ContextLimitModelCacheState = {
   anthropicContext1MEnabled: boolean
   modelContextLimitsCache?: Map<string, number>
@@ -26,10 +19,6 @@ function getAnthropicActualLimit(modelCacheState?: ContextLimitModelCacheState):
     : DEFAULT_ANTHROPIC_ACTUAL_LIMIT
 }
 
-function isAnthropicNoHeaderGaModel(modelID: string): boolean {
-  return ANTHROPIC_NO_HEADER_GA_MODEL_IDS.has(modelID.toLowerCase())
-}
-
 export function resolveActualContextLimit(
   providerID: string,
   modelID: string,
@@ -40,7 +29,7 @@ export function resolveActualContextLimit(
     if (explicit1M === 1_000_000) return explicit1M
 
     const cachedLimit = modelCacheState?.modelContextLimitsCache?.get(`${providerID}/${modelID}`)
-    if (cachedLimit && isAnthropicNoHeaderGaModel(modelID)) return cachedLimit
+    if (cachedLimit) return cachedLimit
 
     return DEFAULT_ANTHROPIC_ACTUAL_LIMIT
   }