shishantbiswas/coder
1
0
Fork 0
mirror of https://github.com/coder/coder.git synced 2026-06-02 20:48:20 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix: normalize command paths to base names in shellparse (#25599)

Browse Source 
Normalize program names in shellparse.Parse to their basename.

Does not rely on filepath.Base because the server may run on either
Linux or Windows where the behavior would differ.

Closes CODAGT-470
This commit is contained in:
Mathias Fredriksson
2026-05-22 13:36:53 +03:00
committed by GitHub
parent 5d40bac79f
commit 0ba702c43f
8 changed files with 187 additions and 159 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/reference/api/chats.md
Generated
+109 -109
View File
@@ -180,115 +180,115 @@ Experimental: this endpoint is subject to change.
Status Code **200**
| Name | Type | Required | Restrictions | Description |
|-----------------------------------|------------------------------------------------------------------------|----------|--------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| `[array item]` | array | false | | |
| `» agent_id` | string(uuid) | false | | |
| `» archived` | boolean | false | | |
| `» build_id` | string(uuid) | false | | |
| `» children` | [codersdk.Chat](schemas.md#codersdkchat) | false | | Children holds child (subagent) chats nested under this root chat. Always initialized to an empty slice so the JSON field is present as []. Child chats cannot create their own subagents, so nesting depth is capped at 1 and this slice is always empty for child chats. |
| `» client_type` | [codersdk.ChatClientType](schemas.md#codersdkchatclienttype) | false | | |
| `» created_at` | string(date-time) | false | | |
| `» diff_status` | [codersdk.ChatDiffStatus](schemas.md#codersdkchatdiffstatus) | false | | |
| `»» additions` | integer | false | | |
| `»» approved` | boolean | false | | |
| `»» author_avatar_url` | string | false | | |
| `»» author_login` | string | false | | |
| `»» base_branch` | string | false | | |
| `»» changed_files` | integer | false | | |
| `»» changes_requested` | boolean | false | | |
| `»» chat_id` | string(uuid) | false | | |
| `»» commits` | integer | false | | |
| `»» deletions` | integer | false | | |
| `»» head_branch` | string | false | | |
| `»» pr_number` | integer | false | | |
| `»» pull_request_draft` | boolean | false | | |
| `»» pull_request_state` | string | false | | |
| `»» pull_request_title` | string | false | | |
| `»» refreshed_at` | string(date-time) | false | | |
| `»» reviewer_count` | integer | false | | |
| `»» stale_at` | string(date-time) | false | | |
| `»» url` | string | false | | |
| `» files` | array | false | | |
| `»» created_at` | string(date-time) | false | | |
| `»» id` | string(uuid) | false | | |
| `»» mime_type` | string | false | | |
| `»» name` | string | false | | |
| `»» organization_id` | string(uuid) | false | | |
| `»» owner_id` | string(uuid) | false | | |
| `» has_unread` | boolean | false | | Has unread is true when assistant messages exist beyond the owner's read cursor, which updates on stream connect and disconnect. |
| `» id` | string(uuid) | false | | |
| `» labels` | object | false | | |
| `»» [any property]` | string | false | | |
| `» last_error` | [codersdk.ChatError](schemas.md#codersdkchaterror) | false | | |
| `»» detail` | string | false | | Detail is optional provider-specific context shown alongside the normalized error message when available. |
| `»» kind` | [codersdk.ChatErrorKind](schemas.md#codersdkchaterrorkind) | false | | Kind classifies the error for consistent client rendering. |
| `»» message` | string | false | | Message is the normalized, user-facing error message. |
| `»» provider` | string | false | | Provider identifies the upstream model provider when known. |
| `»» retryable` | boolean | false | | Retryable reports whether the underlying error is transient. |
| `»» status_code` | integer | false | | Status code is the best-effort upstream HTTP status code. |
| `» last_injected_context` | array | false | | Last injected context holds the most recently persisted injected context parts (AGENTS.md files and skills). It is updated only when context changes, on first workspace attach or agent change. |
| `»» args` | array | false | | |
| `»» args_delta` | string | false | | |
| `»» completed_at` | string(date-time) | false | | Completed at is the time a reasoning part finished streaming, so reasoning duration can be computed as completed_at minus created_at. For interrupted reasoning, this is the interruption time. Absent when reasoning timestamp data was not recorded (e.g. messages persisted before this feature was added). |
| `»» content` | string | false | | The code content from the diff that was commented on. |
| `»» context_file_agent_id` | [uuid.NullUUID](schemas.md#uuidnulluuid) | false | | Context file agent ID is the workspace agent that provided this context file. Used to detect when the agent changes (e.g. workspace rebuilt) so instruction files can be re-persisted with fresh content. |
| `»»» uuid` | string | false | | |
| `»»» valid` | boolean | false | | Valid is true if UUID is not NULL |
| `»» context_file_content` | string | false | | Context file content holds the file content sent to the LLM. Internal only: stripped before API responses to keep payloads small. The backend reads it when building the prompt via partsToMessageParts. |
| `»» context_file_directory` | string | false | | Context file directory is the working directory of the workspace agent. Internal only: same purpose as ContextFileOS. |
| `»» context_file_os` | string | false | | Context file os is the operating system of the workspace agent. Internal only: used during prompt expansion so the LLM knows the OS even on turns where InsertSystem is not called. |
| `»» context_file_path` | string | false | | Context file path is the absolute path of a file loaded into the LLM context (e.g. an AGENTS.md instruction file). |
| `»» context_file_skill_meta_file` | string | false | | Context file skill meta file is the basename of the skill meta file (e.g. "SKILL.md") at the time of persistence. Internal only: restored on subsequent turns so the read_skill tool uses the correct filename even when the agent configured a non-default value. |
| `»» context_file_truncated` | boolean | false | | Context file truncated indicates the file exceeded the 64KiB instruction file limit and was truncated. |
| `»» created_at` | string(date-time) | false | | Created at is the timestamp this part carries. The semantics depend on the part type: for tool-call and tool-result parts it is the time the call was emitted or the result was produced (tool duration is the result's created_at minus the call's created_at); for reasoning parts it is the time reasoning started streaming. |
| `»» data` | array | false | | |
| `»» end_line` | integer | false | | |
| `»» file_id` | [uuid.NullUUID](schemas.md#uuidnulluuid) | false | | |
| `»»» uuid` | string | false | | |
| `»»» valid` | boolean | false | | Valid is true if UUID is not NULL |
| `»» file_name` | string | false | | |
| `»» is_error` | boolean | false | | |
| `»» is_media` | boolean | false | | |
| `»» mcp_server_config_id` | [uuid.NullUUID](schemas.md#uuidnulluuid) | false | | |
| `»»» uuid` | string | false | | |
| `»»» valid` | boolean | false | | Valid is true if UUID is not NULL |
| `»» media_type` | string | false | | |
| `»» name` | string | false | | |
| `»» parsed_commands` | array | false | | Parsed commands holds parsed programs from an execute tool call's shell command, one entry per simple command in source order. Each entry is [program] or [program, arg] where arg is the first non-flag positional argument. Only populated when ToolName is "execute" and the command parses successfully; nil otherwise. |
| `»» provider_executed` | boolean | false | | Provider executed indicates the tool call was executed by the provider (e.g. Anthropic computer use). |
| `»» provider_metadata` | array | false | | Provider metadata holds provider-specific response metadata (e.g. Anthropic cache control hints) as raw JSON. Internal only: stripped by db2sdk before API responses. |
| `»» result` | array | false | | |
| `»» result_delta` | string | false | | |
| `»» result_reset` | boolean | false | | |
| `»» signature` | string | false | | |
| `»» skill_description` | string | false | | Skill description is the short description from the skill's SKILL.md frontmatter. |
| `»» skill_dir` | string | false | | Skill dir is the absolute path to the skill directory inside the workspace filesystem. Internal only: used by read_skill/read_skill_file tools to locate skill files. |
| `»» skill_name` | string | false | | Skill name is the kebab-case name of a discovered skill from the workspace's .agents/skills/ directory. |
| `»» source_id` | string | false | | |
| `»» start_line` | integer | false | | |
| `»» text` | string | false | | |
| `»» title` | string | false | | |
| `»» tool_call_id` | string | false | | |
| `»» tool_name` | string | false | | |
| `»» type` | [codersdk.ChatMessagePartType](schemas.md#codersdkchatmessageparttype) | false | | |
| `»» url` | string | false | | |
| `» last_model_config_id` | string(uuid) | false | | |
| `» last_turn_summary` | string | false | | |
| `» mcp_server_ids` | array | false | | |
| `» organization_id` | string(uuid) | false | | |
| `» owner_id` | string(uuid) | false | | |
| `» owner_name` | string | false | | |
| `» owner_username` | string | false | | |
| `» parent_chat_id` | string(uuid) | false | | |
| `» pin_order` | integer | false | | |
| `» plan_mode` | [codersdk.ChatPlanMode](schemas.md#codersdkchatplanmode) | false | | |
| `» root_chat_id` | string(uuid) | false | | |
| `» status` | [codersdk.ChatStatus](schemas.md#codersdkchatstatus) | false | | |
| `» title` | string | false | | |
| `» updated_at` | string(date-time) | false | | |
| `» warnings` | array | false | | |
| `» workspace_id` | string(uuid) | false | | |
| Name | Type | Required | Restrictions | Description |
|-----------------------------------|------------------------------------------------------------------------|----------|--------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| `[array item]` | array | false | | |
| `» agent_id` | string(uuid) | false | | |
| `» archived` | boolean | false | | |
| `» build_id` | string(uuid) | false | | |
| `» children` | [codersdk.Chat](schemas.md#codersdkchat) | false | | Children holds child (subagent) chats nested under this root chat. Always initialized to an empty slice so the JSON field is present as []. Child chats cannot create their own subagents, so nesting depth is capped at 1 and this slice is always empty for child chats. |
| `» client_type` | [codersdk.ChatClientType](schemas.md#codersdkchatclienttype) | false | | |
| `» created_at` | string(date-time) | false | | |
| `» diff_status` | [codersdk.ChatDiffStatus](schemas.md#codersdkchatdiffstatus) | false | | |
| `»» additions` | integer | false | | |
| `»» approved` | boolean | false | | |
| `»» author_avatar_url` | string | false | | |
| `»» author_login` | string | false | | |
| `»» base_branch` | string | false | | |
| `»» changed_files` | integer | false | | |
| `»» changes_requested` | boolean | false | | |
| `»» chat_id` | string(uuid) | false | | |
| `»» commits` | integer | false | | |
| `»» deletions` | integer | false | | |
| `»» head_branch` | string | false | | |
| `»» pr_number` | integer | false | | |
| `»» pull_request_draft` | boolean | false | | |
| `»» pull_request_state` | string | false | | |
| `»» pull_request_title` | string | false | | |
| `»» refreshed_at` | string(date-time) | false | | |
| `»» reviewer_count` | integer | false | | |
| `»» stale_at` | string(date-time) | false | | |
| `»» url` | string | false | | |
| `» files` | array | false | | |
| `»» created_at` | string(date-time) | false | | |
| `»» id` | string(uuid) | false | | |
| `»» mime_type` | string | false | | |
| `»» name` | string | false | | |
| `»» organization_id` | string(uuid) | false | | |
| `»» owner_id` | string(uuid) | false | | |
| `» has_unread` | boolean | false | | Has unread is true when assistant messages exist beyond the owner's read cursor, which updates on stream connect and disconnect. |
| `» id` | string(uuid) | false | | |
| `» labels` | object | false | | |
| `»» [any property]` | string | false | | |
| `» last_error` | [codersdk.ChatError](schemas.md#codersdkchaterror) | false | | |
| `»» detail` | string | false | | Detail is optional provider-specific context shown alongside the normalized error message when available. |
| `»» kind` | [codersdk.ChatErrorKind](schemas.md#codersdkchaterrorkind) | false | | Kind classifies the error for consistent client rendering. |
| `»» message` | string | false | | Message is the normalized, user-facing error message. |
| `»» provider` | string | false | | Provider identifies the upstream model provider when known. |
| `»» retryable` | boolean | false | | Retryable reports whether the underlying error is transient. |
| `»» status_code` | integer | false | | Status code is the best-effort upstream HTTP status code. |
| `» last_injected_context` | array | false | | Last injected context holds the most recently persisted injected context parts (AGENTS.md files and skills). It is updated only when context changes, on first workspace attach or agent change. |
| `»» args` | array | false | | |
| `»» args_delta` | string | false | | |
| `»» completed_at` | string(date-time) | false | | Completed at is the time a reasoning part finished streaming, so reasoning duration can be computed as completed_at minus created_at. For interrupted reasoning, this is the interruption time. Absent when reasoning timestamp data was not recorded (e.g. messages persisted before this feature was added). |
| `»» content` | string | false | | The code content from the diff that was commented on. |
| `»» context_file_agent_id` | [uuid.NullUUID](schemas.md#uuidnulluuid) | false | | Context file agent ID is the workspace agent that provided this context file. Used to detect when the agent changes (e.g. workspace rebuilt) so instruction files can be re-persisted with fresh content. |
| `»»» uuid` | string | false | | |
| `»»» valid` | boolean | false | | Valid is true if UUID is not NULL |
| `»» context_file_content` | string | false | | Context file content holds the file content sent to the LLM. Internal only: stripped before API responses to keep payloads small. The backend reads it when building the prompt via partsToMessageParts. |
| `»» context_file_directory` | string | false | | Context file directory is the working directory of the workspace agent. Internal only: same purpose as ContextFileOS. |
| `»» context_file_os` | string | false | | Context file os is the operating system of the workspace agent. Internal only: used during prompt expansion so the LLM knows the OS even on turns where InsertSystem is not called. |
| `»» context_file_path` | string | false | | Context file path is the absolute path of a file loaded into the LLM context (e.g. an AGENTS.md instruction file). |
| `»» context_file_skill_meta_file` | string | false | | Context file skill meta file is the basename of the skill meta file (e.g. "SKILL.md") at the time of persistence. Internal only: restored on subsequent turns so the read_skill tool uses the correct filename even when the agent configured a non-default value. |
| `»» context_file_truncated` | boolean | false | | Context file truncated indicates the file exceeded the 64KiB instruction file limit and was truncated. |
| `»» created_at` | string(date-time) | false | | Created at is the timestamp this part carries. The semantics depend on the part type: for tool-call and tool-result parts it is the time the call was emitted or the result was produced (tool duration is the result's created_at minus the call's created_at); for reasoning parts it is the time reasoning started streaming. |
| `»» data` | array | false | | |
| `»» end_line` | integer | false | | |
| `»» file_id` | [uuid.NullUUID](schemas.md#uuidnulluuid) | false | | |
| `»»» uuid` | string | false | | |
| `»»» valid` | boolean | false | | Valid is true if UUID is not NULL |
| `»» file_name` | string | false | | |
| `»» is_error` | boolean | false | | |
| `»» is_media` | boolean | false | | |
| `»» mcp_server_config_id` | [uuid.NullUUID](schemas.md#uuidnulluuid) | false | | |
| `»»» uuid` | string | false | | |
| `»»» valid` | boolean | false | | Valid is true if UUID is not NULL |
| `»» media_type` | string | false | | |
| `»» name` | string | false | | |
| `»» parsed_commands` | array | false | | Parsed commands holds parsed programs from an execute tool call's shell command, one entry per simple command in source order. Each entry is [program] or [program, arg] where arg is the first non-flag positional argument. Program names are normalized to their base name (e.g. /usr/bin/go becomes go). Only populated when ToolName is "execute" and the command parses successfully; nil otherwise. |
| `»» provider_executed` | boolean | false | | Provider executed indicates the tool call was executed by the provider (e.g. Anthropic computer use). |
| `»» provider_metadata` | array | false | | Provider metadata holds provider-specific response metadata (e.g. Anthropic cache control hints) as raw JSON. Internal only: stripped by db2sdk before API responses. |
| `»» result` | array | false | | |
| `»» result_delta` | string | false | | |
| `»» result_reset` | boolean | false | | |
| `»» signature` | string | false | | |
| `»» skill_description` | string | false | | Skill description is the short description from the skill's SKILL.md frontmatter. |
| `»» skill_dir` | string | false | | Skill dir is the absolute path to the skill directory inside the workspace filesystem. Internal only: used by read_skill/read_skill_file tools to locate skill files. |
| `»» skill_name` | string | false | | Skill name is the kebab-case name of a discovered skill from the workspace's .agents/skills/ directory. |
| `»» source_id` | string | false | | |
| `»» start_line` | integer | false | | |
| `»» text` | string | false | | |
| `»» title` | string | false | | |
| `»» tool_call_id` | string | false | | |
| `»» tool_name` | string | false | | |
| `»» type` | [codersdk.ChatMessagePartType](schemas.md#codersdkchatmessageparttype) | false | | |
| `»» url` | string | false | | |
| `» last_model_config_id` | string(uuid) | false | | |
| `» last_turn_summary` | string | false | | |
| `» mcp_server_ids` | array | false | | |
| `» organization_id` | string(uuid) | false | | |
| `» owner_id` | string(uuid) | false | | |
| `» owner_name` | string | false | | |
| `» owner_username` | string | false | | |
| `» parent_chat_id` | string(uuid) | false | | |
| `» pin_order` | integer | false | | |
| `» plan_mode` | [codersdk.ChatPlanMode](schemas.md#codersdkchatplanmode) | false | | |
| `» root_chat_id` | string(uuid) | false | | |
| `» status` | [codersdk.ChatStatus](schemas.md#codersdkchatstatus) | false | | |
| `» title` | string | false | | |
| `» updated_at` | string(date-time) | false | | |
| `» warnings` | array | false | | |
| `» workspace_id` | string(uuid) | false | | |
#### Enumerated Values
docs/reference/api/schemas.md
Generated
+41 -41
View File
@@ -3026,47 +3026,47 @@ AuthorizationObject can represent a "set" of objects, such as: all workspaces in
### Properties
| Name | Type | Required | Restrictions | Description |
|--------------------------------|--------------------------------------------------------------|----------|--------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| `args` | array of integer | false | | |
| `args_delta` | string | false | | |
| `completed_at` | string | false | | Completed at is the time a reasoning part finished streaming, so reasoning duration can be computed as completed_at minus created_at. For interrupted reasoning, this is the interruption time. Absent when reasoning timestamp data was not recorded (e.g. messages persisted before this feature was added). |
| `content` | string | false | | The code content from the diff that was commented on. |
| `context_file_agent_id` | [uuid.NullUUID](#uuidnulluuid) | false | | Context file agent ID is the workspace agent that provided this context file. Used to detect when the agent changes (e.g. workspace rebuilt) so instruction files can be re-persisted with fresh content. |
| `context_file_content` | string | false | | Context file content holds the file content sent to the LLM. Internal only: stripped before API responses to keep payloads small. The backend reads it when building the prompt via partsToMessageParts. |
| `context_file_directory` | string | false | | Context file directory is the working directory of the workspace agent. Internal only: same purpose as ContextFileOS. |
| `context_file_os` | string | false | | Context file os is the operating system of the workspace agent. Internal only: used during prompt expansion so the LLM knows the OS even on turns where InsertSystem is not called. |
| `context_file_path` | string | false | | Context file path is the absolute path of a file loaded into the LLM context (e.g. an AGENTS.md instruction file). |
| `context_file_skill_meta_file` | string | false | | Context file skill meta file is the basename of the skill meta file (e.g. "SKILL.md") at the time of persistence. Internal only: restored on subsequent turns so the read_skill tool uses the correct filename even when the agent configured a non-default value. |
| `context_file_truncated` | boolean | false | | Context file truncated indicates the file exceeded the 64KiB instruction file limit and was truncated. |
| `created_at` | string | false | | Created at is the timestamp this part carries. The semantics depend on the part type: for tool-call and tool-result parts it is the time the call was emitted or the result was produced (tool duration is the result's created_at minus the call's created_at); for reasoning parts it is the time reasoning started streaming. |
| `data` | array of integer | false | | |
| `end_line` | integer | false | | |
| `file_id` | [uuid.NullUUID](#uuidnulluuid) | false | | |
| `file_name` | string | false | | |
| `is_error` | boolean | false | | |
| `is_media` | boolean | false | | |
| `mcp_server_config_id` | [uuid.NullUUID](#uuidnulluuid) | false | | |
| `media_type` | string | false | | |
| `name` | string | false | | |
| `parsed_commands` | array of array | false | | Parsed commands holds parsed programs from an execute tool call's shell command, one entry per simple command in source order. Each entry is [program] or [program, arg] where arg is the first non-flag positional argument. Only populated when ToolName is "execute" and the command parses successfully; nil otherwise. |
| `provider_executed` | boolean | false | | Provider executed indicates the tool call was executed by the provider (e.g. Anthropic computer use). |
| `provider_metadata` | array of integer | false | | Provider metadata holds provider-specific response metadata (e.g. Anthropic cache control hints) as raw JSON. Internal only: stripped by db2sdk before API responses. |
| `result` | array of integer | false | | |
| `result_delta` | string | false | | |
| `result_reset` | boolean | false | | |
| `signature` | string | false | | |
| `skill_description` | string | false | | Skill description is the short description from the skill's SKILL.md frontmatter. |
| `skill_dir` | string | false | | Skill dir is the absolute path to the skill directory inside the workspace filesystem. Internal only: used by read_skill/read_skill_file tools to locate skill files. |
| `skill_name` | string | false | | Skill name is the kebab-case name of a discovered skill from the workspace's .agents/skills/ directory. |
| `source_id` | string | false | | |
| `start_line` | integer | false | | |
| `text` | string | false | | |
| `title` | string | false | | |
| `tool_call_id` | string | false | | |
| `tool_name` | string | false | | |
| `type` | [codersdk.ChatMessagePartType](#codersdkchatmessageparttype) | false | | |
| `url` | string | false | | |
| Name | Type | Required | Restrictions | Description |
|--------------------------------|--------------------------------------------------------------|----------|--------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| `args` | array of integer | false | | |
| `args_delta` | string | false | | |
| `completed_at` | string | false | | Completed at is the time a reasoning part finished streaming, so reasoning duration can be computed as completed_at minus created_at. For interrupted reasoning, this is the interruption time. Absent when reasoning timestamp data was not recorded (e.g. messages persisted before this feature was added). |
| `content` | string | false | | The code content from the diff that was commented on. |
| `context_file_agent_id` | [uuid.NullUUID](#uuidnulluuid) | false | | Context file agent ID is the workspace agent that provided this context file. Used to detect when the agent changes (e.g. workspace rebuilt) so instruction files can be re-persisted with fresh content. |
| `context_file_content` | string | false | | Context file content holds the file content sent to the LLM. Internal only: stripped before API responses to keep payloads small. The backend reads it when building the prompt via partsToMessageParts. |
| `context_file_directory` | string | false | | Context file directory is the working directory of the workspace agent. Internal only: same purpose as ContextFileOS. |
| `context_file_os` | string | false | | Context file os is the operating system of the workspace agent. Internal only: used during prompt expansion so the LLM knows the OS even on turns where InsertSystem is not called. |
| `context_file_path` | string | false | | Context file path is the absolute path of a file loaded into the LLM context (e.g. an AGENTS.md instruction file). |
| `context_file_skill_meta_file` | string | false | | Context file skill meta file is the basename of the skill meta file (e.g. "SKILL.md") at the time of persistence. Internal only: restored on subsequent turns so the read_skill tool uses the correct filename even when the agent configured a non-default value. |
| `context_file_truncated` | boolean | false | | Context file truncated indicates the file exceeded the 64KiB instruction file limit and was truncated. |
| `created_at` | string | false | | Created at is the timestamp this part carries. The semantics depend on the part type: for tool-call and tool-result parts it is the time the call was emitted or the result was produced (tool duration is the result's created_at minus the call's created_at); for reasoning parts it is the time reasoning started streaming. |
| `data` | array of integer | false | | |
| `end_line` | integer | false | | |
| `file_id` | [uuid.NullUUID](#uuidnulluuid) | false | | |
| `file_name` | string | false | | |
| `is_error` | boolean | false | | |
| `is_media` | boolean | false | | |
| `mcp_server_config_id` | [uuid.NullUUID](#uuidnulluuid) | false | | |
| `media_type` | string | false | | |
| `name` | string | false | | |
| `parsed_commands` | array of array | false | | Parsed commands holds parsed programs from an execute tool call's shell command, one entry per simple command in source order. Each entry is [program] or [program, arg] where arg is the first non-flag positional argument. Program names are normalized to their base name (e.g. /usr/bin/go becomes go). Only populated when ToolName is "execute" and the command parses successfully; nil otherwise. |
| `provider_executed` | boolean | false | | Provider executed indicates the tool call was executed by the provider (e.g. Anthropic computer use). |
| `provider_metadata` | array of integer | false | | Provider metadata holds provider-specific response metadata (e.g. Anthropic cache control hints) as raw JSON. Internal only: stripped by db2sdk before API responses. |
| `result` | array of integer | false | | |
| `result_delta` | string | false | | |
| `result_reset` | boolean | false | | |
| `signature` | string | false | | |
| `skill_description` | string | false | | Skill description is the short description from the skill's SKILL.md frontmatter. |
| `skill_dir` | string | false | | Skill dir is the absolute path to the skill directory inside the workspace filesystem. Internal only: used by read_skill/read_skill_file tools to locate skill files. |
| `skill_name` | string | false | | Skill name is the kebab-case name of a discovered skill from the workspace's .agents/skills/ directory. |
| `source_id` | string | false | | |
| `start_line` | integer | false | | |
| `text` | string | false | | |
| `title` | string | false | | |
| `tool_call_id` | string | false | | |
| `tool_name` | string | false | | |
| `type` | [codersdk.ChatMessagePartType](#codersdkchatmessageparttype) | false | | |
| `url` | string | false | | |
## codersdk.ChatMessagePartType