- Implemented in-house, zero-dependency utility modules in `pi-utils` covering DOM manipulation, markdown parsing, templating, browser automation helpers, and terminal buffers.
- Migrated packages across the repository to consume the new internal utilities and `omptype` schema validators instead of external dependencies.
- Removed multiple external runtime and development dependencies including Zod, Marked, LRU cache, Turndown, and Puppeteer browser packages.
Object-form additionalProperties/unevaluatedProperties are genuine
subschemas whose interior may still hold the bare booleans emitted by
toolWireSchema; skipping them let llama.cpp reach an Unrecognized
schema: true. Boolean forms stay intact for closed/open-object grammar
semantics. Flagged by Codex review on #5919.
Local grammar-constrained OpenAI-compatible servers (llama.cpp, LM Studio,
vLLM) build a GBNF grammar from tool JSON Schemas and 400 with
"Unrecognized schema: true" on a bare boolean subschema. The task tool's
open outputSchema field normalizes to boolean `true` (issue #1179), which
the converter cannot compile, breaking every tool request.
Add a "grammar" tool-schema flavor, auto-detected for local backends, that
widens bare boolean/empty subschemas into a value-accepting primitive union
while preserving closed-object `additionalProperties: false`.
Fixes#5914
Added not to the CCA residual incompatibility gate so false boolean subschemas fall back before reaching the legacy object-shaped parameters wire.
Made residual scanning schema-map aware to avoid treating a property literally named not as the unsupported keyword, and added root, property, and dependentSchemas regressions.
Fixes#5604
Generalized the schema-map walk context beyond properties to include patternProperties, dependencies, dependentSchemas, $defs, and definitions. Each arbitrary map entry is now treated as a subschema, so bare booleans coerce before reaching Google/CCA.
Reused the shared map/array keyword tables in the Ollama sanitizer and added a dependentSchemas regression for both Google and CCA normalizers.
Fixes#5604
Boolean JSON Schema subschemas (`true`/`false`, draft 6+) in MCP tool
inputs passed through normalizeSchemaForGoogle/normalizeSchemaForCCA
untouched. The Cloud Code Assist / Gemini protobuf Schema type has no
representation for a bare boolean, so requests bounced with a 400
INVALID_ARGUMENT before reaching the model.
Coerce booleans to their object equivalents (`true` -> `{}`,
`false` -> `{ not: {} }`) at the single normalizeSchemaNode choke point,
but only in genuine subschema slots (root, combiner branches,
subschema-valued keywords, property values). Keyword-slot booleans
(`nullable`, `enum` entries, `additionalProperties`) stay untouched so
Moonshot/MCP open-record markers survive.
Fixes#5604
- Wrapped generated multi-type branches in allOf so existing sibling anyOf constraints remain conjunctive.
- Added a regression case for schemas that combine type arrays with their own anyOf.
Refs #4488
- Replaced boolean true / empty subschemas with an anyOf union over every primitive JSON type so grammar-constrained samplers (llama.cpp) keep advertising "any JSON value" for unconstrained fields.
- Dropped the WeakMap identity cache; the sanitizer no longer relies on module-external state.
- Updated the provider regression test and changelog entry to the widened shape.
Refs #4488
- Added an Ollama-specific tool schema sanitizer for boolean subschemas, boolean additionalProperties/unevaluatedProperties, and nullable type arrays.
- Applied the sanitizer in the native Ollama chat tool serializer and covered the provider payload contract with a regression test.
Fixes#4488
- Migrated 288 lines of scattered error classification logic from `utils/error-id.ts` into a cohesive `packages/ai/src/error/` module with 13 specialized submodules covering flags, classes, OAuth, providers, rate-limiting, and finalization.
- Replaced 100+ generic `Error` throws across 60+ provider and registry files with semantic `AIError.*` classes (e.g., `AIError.MissingApiKeyError`, `AIError.OAuthError`, `AIError.ProviderResponseError`), improving error diagnostics and retry logic.
- Consolidated error utility imports from `pi-utils` and scattered classification functions into a single `AIError` namespace, reducing coupling and simplifying error handling across all packages.
- Integrated comprehensive loop guard support for DeepSeek and assistant prose patterns, including configurable stream checks.
- Implemented Moonshot Flavored JSON Schema (MFJS) normalization for improved tool compatibility and enum type inference.
- Added support for Ollama reasoning effort backfilling and Grok-specific service tier cost tracking across providers.
- Expanded model catalog with new entries and unified compatibility logic for improved OpenRouter API integration.
- Added native support for parsing, normalized validation, and serialization of ArkType schemas throughout the agent pipeline.
- Implemented pruning of unconstrained union branches and normalization helpers to handle unrepresentable strict-mode branches.
- Patched ArkType's schema package to preserve declared object key order during serialization.
- Integrated ArkType schemas into `agent-loop` and migrated coding agent tool parameters to ArkType format.
Rewrote unsupported lookaround patternProperties keys to a supported catch-all pattern instead of dropping their value schemas, preserving dynamic-key tool arguments when additionalProperties is false.
Extended sanitizer and Codex conversion regression coverage for the closed dynamic-key case.
Fixes#2784
Converted schema nodes emptied by OpenAI Responses lookaround stripping to boolean true so pattern-only nodes keep the existing empty-schema semantics.
Extended sanitizer and Codex regression coverage for pattern-only property and propertyNames schemas.
Fixes#2784
Removed JSON Schema pattern values containing regex lookaround from OpenAI Responses/Codex tool schemas so incompatible MCP tools do not poison the request.
Added schema-normalization and Codex conversion regression coverage for Figma-style fileKey patterns.
Fixes#2784
- Tracked whether strict tool schemas were actually applied and used that flag when deciding strict-to-nonstrict retries.
- Recorded strict-tool failures in session state so later requests skip strict mode retries and avoid extra error rounds.
- Hardened strict schema normalization by flattening nested pure anyOf unions and expanded schema error matching for additional invalid-schema rejections.
Only flatten optional anyOf schemas when the union wrapper has no sibling constraints that would exclude null.
Added a strict-schema regression for constrained anyOf properties so null remains an outer branch.
Flattened optional union tool schemas during strict enforcement so OpenRouter DeepSeek V4 no longer receives nested anyOf branches without a type.
Added schema-level and OpenRouter DeepSeek payload regressions for optional string-or-array tool parameters.
Fixes#2270
in-band Gemini errors, promptFeedback blocks, and missing finishReason no longer report success; toolUse override stops masking SAFETY/MALFORMED finishes; schema normalization keeps DAG-shared subtrees while detecting true cycles; Google/AWS shared credential resolution detached from first caller's signal and bounded by own timeout; Bedrock keeps toolConfig under toolChoice none; eventstream cancels body on abnormal exit.
Address review feedback:
- Replace `as string` assertion with typed `chosenType` local
- Strip sibling keys from nextSchema that were copied via
copySchemaWithout but belong to a type other than the chosen one
(e.g. sibling `items` on a now-string-typed schema)
- Export ALL_CCA_TYPE_SPECIFIC_KEYS from fields.ts for sibling filtering
- Add regression test for the sibling-key edge case
When collapseMixedTypeCombinerVariants collapses an anyOf with mixed
types (e.g. string | array), it previously picked the first non-null
type but indiscriminately copied ALL mergedVariantFields — including
type-specific keys like "items" that only belong to array. This
produced schemas like {type: "string", items: {...}} which Google
Cloud Code Assist API rejects with 400.
Fix: filter mergedVariantFields against the chosen types allowed keys
(CLOUD_CODE_ASSIST_TYPE_SPECIFIC_KEYS) before copying, so array-only
keys are dropped when the winner is string (and vice versa).
Fixes 400 error on github tools "pr" parameter (anyOf string/array).
- Updated `normalizeSchemaNode` to keep `insideProperties` active only when entering an outer `properties` map, not when the key appears within an existing properties object.
- Added a regression test for `google-tool-schema` ensuring `propertyNames` and `additionalProperties` are removed from a schema property literally named `properties`.
Move {}→true normalization out of zodToWireSchema/postProcess (Zod-only)
into a dedicated, exported normalizeEmptySchemas helper that toolWireSchema
calls for both the Zod and TypeBox/raw-JSON-Schema branches. Every provider
that calls toolWireSchema (OpenAI, Anthropic, Google, Ollama, Bedrock,
Cursor) now receives the normalized schema regardless of how the tool was
authored.
Verified strict-mode opt-out across providers:
- OpenAI: hasUnrepresentableStrictObjectMap hits === true branch (was
isJsonObject({}) branch), same result -> strict: false.
- Anthropic: normalizeAnthropicStrictSchemaNode opts out via
additionalProperties !== false (still true for true) -> strict: false.
- Google: normalizeSchemaForGoogle strips additionalProperties regardless
(UNSUPPORTED_SCHEMA_FIELDS, pre-existing behavior).
The normalizeOpenAIResponsesSchemaNode guard stays as a safety net for
callers that invoke sanitizeSchemaForOpenAIResponses directly on schemas
that bypass the wire-schema pipeline (e.g. fixtures, debug paths).
Adds cross-provider tests covering both Zod and TypeBox inputs and verifying
downstream provider behavior with the normalized form.
Fixes#1179
z.unknown() emits {} (empty JSON Schema) as the value for additionalProperties,
items, and other schema-valued positions. This is semantically equivalent to
boolean true per JSON Schema draft 2020-12 §4.3.1, but grammar-constrained
samplers (llama.cpp, etc.) interpret the object form as "generate an empty
object" rather than "any JSON value". Models using such samplers therefore
emit {} for every open-typed field, including extra.title in plan-mode resolve
calls (issue #1179). This affected all tools with open schemas, not just the
resolve tool -- MCP tools and any tool using z.unknown() in value positions
have the same problem.
Fix: normalize any {} in schema-valued positions to true in two places:
- zodToWireSchema postProcess/walk: covers Zod-based tools across all providers
- normalizeOpenAIResponsesSchemaNode: covers all tools (Zod, TypeBox, MCP)
going to openai-responses-compatible endpoints (llama.cpp, Azure Responses)
Adds isJsonObjectEmpty() helper to schema/types.ts (for...in based, no allocation).
Fixes#1179
Rewrites JSON-roundtripped Zod 4 schema objects that leak Zod internals as
JSON Schema keywords (e.g., `type:"enum"`, `enum:{...}`) into valid
JSON Schema 2020-12. This prevents validation failures when such schemas
are used as tool input schemas (e.g., from MCP servers).
Updates `isZodSchema` to reject deserialized Zod impostors that retain
`_zod` but lose their prototype.
Wires `toJSON` methods onto TypeBox shim schemas to ensure `JSON.stringify`
produces clean JSON Schema, preventing future leaks.
Fixes#1101
Added dependencies and contentSchema to the OpenAI Responses schema-position sets so draft-04..07 dependencies maps and draft 2019-09 contentSchema nodes also get the properties-on-object normalization. Recognized array-form type declarations that include object. Preserved malformed non-array oneOf payloads instead of dropping them. Documented the cycle-safe cache seed and added a self-referential regression test.
Refs #1147
Normalized OpenAI Responses schema sanitization so object schema nodes always include properties while only traversing schema-valued positions. Wired the same sanitizer into the OpenAI Codex Responses tool conversion path and added regression coverage for no-argument MCP-style schemas and literal payload preservation.
Fixes#1147
- Extracted `description` from type-array and nullable branches so it lives on the wrapper, not duplicated onto each variant.
- Replaced inline enum-type inference with `inferStrictPrimitiveTypeFromEnumOrConst`, covering both `enum` and `const` in sanitize and enforce paths.
- Mixed-primitive enums and non-primitive consts now fall back to non-strict instead of producing a typeless schema that OpenAI rejects on the wire.
- Moved sanitizeSchemaForStrictMode, enforceStrictSchema, and tryEnforceStrictSchema into normalize.ts.
- Moved sanitizeSchemaForOpenAIResponses/rewriteOneOfToAnyOf into normalize.ts alongside other normalizers.
- Removed strict-mode.ts and its public re-export; adapt.ts now only exposes NO_STRICT and adaptSchemaForStrict.
- Dropped StringEnum helper from strict-mode.ts (already removed from public API).
- Implemented a unified normalization flow by switching Google/CCA handling to normalizeSchemaForGoogle/CCA.
- Added normalize.ts with recursive node normalization, nullable-union checks, and combiner collapsing.
- Removed sanitize-google.ts and normalize-cca.ts, replacing them with normalize exports in schema indexes.
- Added spill-to-description utilities with spill/paren modes and `$defs` exclusion for unsupported fields.
- Updated MCP bridge and schema tests to use normalizeSchemaFor* APIs with expanded compatibility checks.
- Documented normalization behavior changes and breaking rename in constraints and package changelog files.
can1357