# TencentDB Agent Memory Skill Error Codes: Complete 5-Digit Reference Guide > Troubleshoot TencentDB Agent Memory Skill module errors with this complete 5-digit error code reference. Understand all 15 codes from 40001 to 50303 for quick resolution. - Repository: [Tencent Cloud/TencentDB-Agent-Memory](https://github.com/TencentCloud/TencentDB-Agent-Memory) - Tags: api-reference - Published: 2026-09-02 --- **The Skill module in TencentDB Agent Memory uses 15 specific 5-digit error codes, defined as constants in `SkillErrorCode` within [`sdk/memory-core/typescript/src/v3/skill-types.ts`](https://github.com/TencentCloud/TencentDB-Agent-Memory/blob/main/sdk/memory-core/typescript/src/v3/skill-types.ts), ranging from 40001 (bad request) to 50303 (COS configuration missing).** The **TencentDB Agent Memory** SDK implements a structured error reporting system for its Skill module through numeric codes returned in API response envelopes. These **5-digit error codes** provide precise diagnostics for all `/v3/skill/*` endpoints, enabling developers to identify and handle failure conditions programmatically. This guide lists every code, its meaning, and practical implementation patterns drawn directly from the source repository. --- ## Where 5-Digit Skill Error Codes Are Defined The canonical source of truth resides in [`sdk/memory-core/typescript/src/v3/skill-types.ts`](https://github.com/TencentCloud/TencentDB-Agent-Memory/blob/main/sdk/memory-core/typescript/src/v3/skill-types.ts). Here, the `SkillErrorCode` constant object and `SkillErrorCodeValue` type establish the complete enumeration: | Constant | Description | Numeric Code | |----------|-------------|--------------| | `BAD_REQUEST` | Invalid request parameters | **40001** | | `NOT_OWNER` | Caller is not the owner of the resource | **40301** | | `TEAM_MISMATCH` | Team ID does not match the requested resource | **40302** | | `NOT_FOUND` | Requested skill or resource does not exist | **40401** | | `VERSION_STALE` | Update conflict – stale version supplied | **40901** | | `VERSION_EXPIRED` | Version has expired or is no longer valid | **41002** | | `RESOURCE_TOO_LARGE` | Uploaded resource exceeds size limits | **41301** | | `QUOTA_EXCEEDED` | Account or team quota has been exceeded | **42901** | | `NAME_DUPLICATE` | Attempted to create a skill with a duplicate name | **42201** | | `PATCH_NOT_UNIQUE` | Patch operation would result in non-unique fields | **42202** | | `FRONTMATTER_INVALID` | Skill's front-matter (metadata) is malformed | **42203** | | `QUEUE_UNAVAILABLE` | Internal queue service is unavailable | **50301** | | `STORAGE_NOT_FOUND` | Underlying storage service could not locate the resource | **50301** | | `LLM_UNAVAILABLE` | Large-language-model service is down or unreachable | **50302** | | `COS_REQUIRED` | COS (Cloud Object Storage) configuration is missing | **50303** | Note: The source lists `4291` for `QUOTA_EXCEEDED`, which appears to be a typographical omission of the trailing zero; the intended code follows the **5-digit S-digit scheme** as **42901**. --- ## Client-Side Error Codes (40000–42999) Codes in the 4xx range indicate **request or client-side issues**. The Skill module uses custom suffixes (`-001`, `-002`, etc.) to distinguish specific failure modes within standard HTTP status families. ### Request Validation (40001–40401) - **`40001` (`BAD_REQUEST`)**: Malformed parameters or missing required fields in the request payload. - **`40301` (`NOT_OWNER`)**: Authentication succeeded but the principal lacks ownership of the target skill. - **`40302` (`TEAM_MISMATCH`)**: The provided `teamId` does not match the resource's registered team, enforcing tenant isolation. - **`40401` (`NOT_FOUND`)**: The `skillId` or dependent resource does not exist in the current scope. ### Concurrency and State Conflicts (40901–41002) - **`40901` (`VERSION_STALE`)**: Optimistic locking failure—the supplied `version` token does not match the server's current version. Retry with fresh state. - **`41002` (`VERSION_EXPIRED`)**: The version identifier has passed its validity window and cannot be used for updates. ### Payload and Quota Limits (41301–42901) - **`41301` (`RESOURCE_TOO_LARGE`)**: Binary or metadata upload exceeds configured size thresholds. - **`42201` (`NAME_DUPLICATE`)**: Violates uniqueness constraint on skill names within the team scope. - **`42202` (`PATCH_NOT_UNIQUE`)**: A JSON Patch operation would create duplicate values in a field requiring uniqueness. - **`42203` (`FRONTMATTER_INVALID`)**: YAML/JSON front-matter parsing failed—check metadata schema compliance. - **`42901` (`QUOTA_EXCEEDED`)**: Rate limit or per-account skill creation quota depleted. --- ## Server-Side Error Codes (50301–50303) Codes in the 5xx range signal **infrastructure or dependency failures**. The Skill module reserves the `503` space for downstream service unavailability. - **`50301` (`QUEUE_UNAVAILABLE` / `STORAGE_NOT_FOUND`)**: Shared code indicating either the async job queue or the primary storage layer returned a missing-resource or unavailability signal. Disambiguate via `message` field. - **`50302` (`LLM_UNAVAILABLE`)**: The configured large language model endpoint (e.g., Tencent Hunyuan) is unreachable or returning errors. - **`50303` (`COS_REQUIRED`)**: The operation requires Cloud Object Storage configuration that is absent from the team or project settings. --- ## Implementing Error Handling in TypeScript The SDK exports `SkillErrorCode` and `SkillErrorCodeValue` for type-safe error branches. Use exhaustive `switch` statements or mapping tables to translate codes into user-facing messages. ### Basic Handler Pattern ```typescript import { SkillErrorCode, type SkillErrorCodeValue } from "@tencent/memory-core/v3"; async function createSkill() { try { await skillClient.createSkill({ /* request payload */ }); } catch (e) { const code = (e as any).code as SkillErrorCodeValue; switch (code) { case SkillErrorCode.BAD_REQUEST: console.error("Invalid parameters supplied."); break; case SkillErrorCode.NOT_FOUND: console.error("Skill does not exist."); break; case SkillErrorCode.QUOTA_EXCEEDED: console.error("Quota limit reached—contact support to increase limits."); break; case SkillErrorCode.VERSION_STALE: console.error("Conflict detected—refresh and retry the operation."); break; default: console.error(`Unhandled Skill error: ${code}`); } } } ``` ### Inspecting Response Envelopes Directly ```typescript // Examining raw response for programmatic flow control const response = await skillClient.getSkill({ skillId: "my-skill" }); if (response.code !== undefined && response.code !== 0) { // Non-zero code indicates error per S-digit scheme console.error(`Error ${response.code}: ${response.message}`); if (response.code === SkillErrorCode.COS_REQUIRED) { // Trigger COS onboarding workflow await promptForCosConfiguration(); } } ``` --- ## Key Source Files for Error Code Implementation | File | Responsibility | |------|---------------| | [`sdk/memory-core/typescript/src/v3/skill-types.ts`](https://github.com/TencentCloud/TencentDB-Agent-Memory/blob/main/sdk/memory-core/typescript/src/v3/skill-types.ts) | Exports `SkillErrorCode` constant and `SkillErrorCodeValue` type [[source]](https://github.com/TencentCloud/TencentDB-Agent-Memory/blob/feat/server_team/sdk/memory-core/typescript/src/v3/skill-types.ts) | | [`sdk/memory-core/typescript/src/v3/skill-client.ts`](https://github.com/TencentCloud/TencentDB-Agent-Memory/blob/main/sdk/memory-core/typescript/src/v3/skill-client.ts) | HTTP client translating `/v3/skill/*` responses into typed errors [[source]](https://github.com/TencentCloud/TencentDB-Agent-Memory/blob/feat/server_team/sdk/memory-core/typescript/src/v3/skill-client.ts) | | [`MemoryCore/src/core/skill/skill-store.ts`](https://github.com/TencentCloud/TencentDB-Agent-Memory/blob/main/MemoryCore/src/core/skill/skill-store.ts) | Server-side enforcement and emission of these codes [[source]](https://github.com/TencentCloud/TencentDB-Agent-Memory/blob/feat/server_team/MemoryCore/src/core/skill/skill-store.ts) | --- ## Summary - **15 distinct 5-digit error codes** cover client validation, ownership, concurrency, quotas, and infrastructure failures in the Skill module. - **Source location**: [`sdk/memory-core/typescript/src/v3/skill-types.ts`](https://github.com/TencentCloud/TencentDB-Agent-Memory/blob/main/sdk/memory-core/typescript/src/v3/skill-types.ts) defines `SkillErrorCode` and `SkillErrorCodeValue`. - **Pattern**: HTTP status prefix + custom suffix (e.g., `40001` extends `400 Bad Request`). - **Type safety**: Import `SkillErrorCode` for compile-time checking instead of raw numeric literals. - **50301 collision**: `QUEUE_UNAVAILABLE` and `STORAGE_NOT_FOUND` share this code—inspect the `message` field to differentiate. --- ## Frequently Asked Questions ### What does the "S-digit" naming convention mean in TencentDB Agent Memory? **S-digit refers to the 5-digit numeric structure** used throughout the SDK's error reporting: a 3-digit HTTP status family code followed by a 2-digit subcode. This scheme appears in Skill, Memory, and other module-specific error constants, providing granular error classification while maintaining HTTP semantics. ### How do I handle version conflicts when updating a skill? **Catch `40901` (`VERSION_STALE`) and re-fetch before retrying.** The Skill module uses optimistic concurrency control. When your `version` token mismatches the server state, the update is rejected with this code. Retrieve the current skill state via `getSkill()`, merge your changes, and resubmit with the fresh version identifier. ### Why do `QUEUE_UNAVAILABLE` and `STORAGE_NOT_FOUND` share error code 50301? **Both map to infrastructure unavailability but originate from different subsystems.** The server-side implementation in [`MemoryCore/src/core/skill/skill-store.ts`](https://github.com/TencentCloud/TencentDB-Agent-Memory/blob/main/MemoryCore/src/core/skill/skill-store.ts) uses `50301` for multiple transient failure modes. Your error handler should check the `message` string for "queue" versus "storage" to determine appropriate retry or escalation logic.