Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@amendor-mcp show me the change requests"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Connect your coding agent to Amendor.
Amendor lets the people you build for request changes right on your live site. Requests land on a board. This connector pulls them into your agent (Claude Code, Cursor, Cline, Codex, any MCP client) so it can build each one on a branch and open a pull request.
Sign in at amendor.site and create a project.
Go to Settings and copy the connector command. It already has your token filled in and looks like this:
claude mcp add amendor --env AMENDOR_API=https://amendor.site --env AMENDOR_TOKEN=your-token -- npx -y amendor-mcpRun it in your repo.
AMENDOR_TOKEN is the key that ties the connector to your Amendor account. You do not make it up. Amendor generates it and shows it inside the ready-made command on the Settings page, so create a project first, then copy the whole command.
Related MCP server: bgagents
Ask your agent: "show me the change requests," pick one, "build that one." It opens a pull request, your host builds a preview, the requester approves, you merge.
Tools: list_change_requests, get_change_request, start_build, set_preview_url, publish_preview, update_status, ship_change.
MIT license. More at amendor.site.
Get full detail for one change request so you can implement it: the user request, the exact element they pointed at (CSS selector + outerHTML), the page URL, and a screenshot URL. Call this before making the change.
| Name | Required | Description | Default |
|---|---|---|---|
| id | Yes | Change request id, from list_change_requests. |
| Name | Required | Description |
|---|---|---|
| id | Yes | Change request id. |
| status | Yes | Current status. |
| page_url | Yes | URL of the page the request was made on. |
| requested_by | Yes | Requester email, or "anonymous". |
| request_text | Yes | What the user wants. |
| revision_requested | Yes | True if the requester reviewed a preview and asked for changes; push to the same branch/PR. |
| revision_comment | Yes | The revision note, or empty string if none. |
| elements | Yes | Elements attached to the request; empty for a general request. |
| screenshot_url | Yes | Absolute screenshot URL, or empty string. |
| attachment_urls | Yes | Absolute URLs of any extra attachments. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=true and destructiveHint=false, establishing safety. The description adds value by detailing the exact return fields (user request, element, page URL, screenshot URL), which is beyond what annotations provide. No contradiction; the description complements annotations with useful behavioral context about the response content.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single sentence that immediately states the tool's purpose ('Get full detail for one change request') and then lists key details. Every word is necessary; there is no redundancy or filler. The instruction to call before making the change is efficiently integrated.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool's simplicity (1 param, output schema exists, annotations cover safety), the description is complete. It explains what the tool returns, when to use it, and how to obtain the required input. The presence of an output schema means detailed return value documentation is not needed in the description.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The single parameter 'id' is fully described in the schema (100% coverage) as 'Change request id, from list_change_requests.' The description does not add additional semantic meaning beyond what the schema already provides. Baseline 3 is appropriate since the schema handles the explanation.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the verb 'Get', the resource 'full detail for one change request', and specifies what is included (user request, element, page URL, screenshot URL). It also implies the tool's role in the workflow by saying 'Call this before making the change', distinguishing it from siblings like list_change_requests (list all) and update_status (update).
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description explicitly says 'Call this before making the change', providing clear when-to-use guidance. While it doesn't list exclusions or alternatives, the context of sibling tools (list, set, update) makes the intended usage clear. The instruction to use before implementation is strong guidance.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
List change requests submitted by end users on sites you ship. Optionally filter by status. Start here to see what people are asking for.
| Name | Required | Description | Default |
|---|---|---|---|
| status | No | Optional status filter: new, accepted, building, preview, approved, shipped, or rejected. Omit to list every open request. |
| Name | Required | Description |
|---|---|---|
| requests | Yes | Matching change requests, newest first. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint, idempotentHint, openWorldHint, and destructiveHint false. The description adds no behavioral traits beyond these, such as pagination or ordering. It does not contradict annotations.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two sentences with no extraneous words: first sentence states purpose and filter, second provides guidance. Front-loaded and efficient.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the simple tool (one optional param, output schema present), the description covers the core functionality. It omits details like result ordering or limits, but these are not critical for a straightforward list tool.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100% with a full description of the status parameter and its values. The description only echoes 'Optionally filter by status,' adding no new meaning.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool lists change requests submitted by end users, with optional status filtering. It distinguishes from sibling tools like get_change_request (single request) and update_status (status update).
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description advises 'Start here to see what people are asking for,' implying it's the entry point for viewing requests. It does not explicitly contrast with alternatives or mention when not to use, but the sibling tool names provide context.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Mark a change request ready for the requester to review. The preview is already live from the worktree; this flips status to "in preview" and surfaces the link.
| Name | Required | Description | Default |
|---|---|---|---|
| id | Yes | Change request id whose local preview is ready. |
| Name | Required | Description |
|---|---|---|
| id | Yes | Change request id. |
| status | Yes | New status, set to "preview". |
| review_url | Yes | Page where the requester reviews and approves the change. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Adds significant context beyond annotations: explains that the preview is already live, this flips status and surfaces a link. Annotations already indicate idempotent and non-destructive, description enriches with specific behavior.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two concise sentences: first states purpose, second provides prerequisite and effect. No wasted words.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given simple tool with one param, output schema present, and annotations covering safety, description fully explains purpose, behavior, and result. Complete for agent usage.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema has 100% coverage with description for the single parameter 'id'. Description does not add extra meaning beyond the schema, baseline of 3 appropriate.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Clearly states the action 'Mark a change request ready for the requester to review' and specifies the effect (flips status to 'in preview' and surfaces link). Distinguishes from siblings like set_preview_url and ship_change by noting the preview is already live.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Implies usage after preview is live from worktree but does not explicitly state when not to use or provide alternatives. Clear context for a simple state transition.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Attach a deploy-preview URL to a change request and mark it ready for the requester to review.
| Name | Required | Description | Default |
|---|---|---|---|
| id | Yes | Change request id. | |
| url | Yes | Deploy-preview URL (from your host) that the requester will review. |
| Name | Required | Description |
|---|---|---|
| id | Yes | Change request id. |
| status | Yes | New status, set to "preview". |
| review_url | Yes | Page where the requester reviews and approves the change. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations indicate idempotent and non-destructive, and description mentions 'mark it ready' as behavioral effect. However, it does not clarify behavior if URL already exists or if change request is not found. Adds some value beyond annotations but incomplete.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Single sentence conveying all necessary information without extraneous words. Well-structured and easy to parse.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a simple tool with two parameters and output schema present, description covers core purpose and effect. However, missing details like prerequisite (change request must exist) or error cases, mildly reducing completeness.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100%, so parameters are documented. Description repeats 'deploy-preview URL' but adds no new semantics beyond the schema, such as constraints on URL format or implications of the id. Baseline score applies.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Description clearly states the action: attaching a deploy-preview URL to a change request and marking it ready for review. It uses specific verb and resources, differentiating from sibling tools like get_change_request (read) and update_status (status change without URL).
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
No explicit guidance on when to use this tool vs alternatives like update_status. It implies use when attaching a URL, but lacks exclusions or context for when not to use it. Adequate but minimal.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Ship an approved change: merge the task branch into the project base, remove the worktree, and mark it shipped. Production (the live site) now reflects the change.
| Name | Required | Description | Default |
|---|---|---|---|
| id | Yes | Change request id to ship (should be approved first). |
| Name | Required | Description |
|---|---|---|
| id | Yes | Change request id. |
| merged | Yes | Branch that was merged. |
| into | Yes | Base branch it was merged into. |
| status | Yes | New status, set to "shipped". |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
The description details destructive actions (merge, remove worktree, mark shipped) and states the production effect, going beyond annotations which only provide destructiveHint=true. It adds context about the change's lifecycle but does not specify error cases (e.g., if change not approved).
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is two short sentences, front-loading the main action and providing immediate clarity. Every part is essential with no redundancy or filler.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool has a single parameter and an output schema (indicated but not shown), the description sufficiently covers the tool's behavior: what it does, what it affects, and the prerequisite. There are no missing gaps for a simple mutation tool.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The schema already fully describes the single parameter 'id' with the same text as the description ('Change request id to ship (should be approved first).'). The tool description adds no new meaning beyond the schema, resulting in a baseline score of 3.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description explicitly states the tool's action: 'Ship an approved change: merge the task branch into the project base, remove the worktree, and mark it shipped.' It clearly identifies the resource (change request) and the expected outcome (production reflects the change), differentiating it from sibling tools like get_change_request or list_change_requests.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description indicates the prerequisite that the change must be approved ('Ship an approved change'), guiding when to use it. It does not explicitly list when not to use or alternatives, but the prerequisite and action make usage clear in context.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
LOCAL MODE ONLY (repo on this machine, no host). Cuts a branch + git worktree the local Amendor serves at /preview//. For the normal GitHub flow, do NOT call this: instead branch "relay/task-" in your own repo clone, edit, commit, push, and open a PR tagged "relay-task:" — the host builds the preview and the webhook captures it automatically.
| Name | Required | Description | Default |
|---|---|---|---|
| id | Yes | Change request id to start a local build for. |
| Name | Required | Description |
|---|---|---|
| id | Yes | Change request id. |
| branch | Yes | Branch that was created. |
| base | Yes | Base branch it was cut from. |
| worktree | Yes | Local folder to edit; commit your change here. |
| preview_url | Yes | URL that serves the worktree live for review. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
The description adds behavioral context beyond annotations: it specifies that the tool modifies local repo state (cuts branch and worktree) and serves a preview at a specific path. It lacks details on potential side effects (e.g., whether it overwrites existing worktrees) but provides useful behavioral scope.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is concise, front-loads the critical note about local-only mode, and separates the alternative flow. It could be slightly more structured (e.g., bullet points) but is efficient and clear.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool's simplicity (1 param, output schema exists), the description covers purpose, usage, and behavioral context adequately. It doesn't explain prerequisites (e.g., Amendor running) or error cases, but these are minor given the clarity.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100%; the description reinforces the parameter 'id' by showing its use in the preview path (/preview/<id>/), adding meaning beyond the schema's description. It does not add constraints like allowed formats, but is adequate.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states that the tool starts a local build by cutting a branch and setting up a git worktree served at /preview/<id>/. It explicitly distinguishes from the normal GitHub flow, making the purpose unambiguous.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description explicitly states when to use the tool (LOCAL MODE ONLY) and when not to (normal GitHub flow), providing a detailed alternative workflow. This gives clear guidance on selection vs. siblings.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Update a change request status as it moves through the build and review flow.
| Name | Required | Description | Default |
|---|---|---|---|
| id | Yes | Change request id. | |
| status | Yes | New status: new, accepted, building, preview, approved, shipped, or rejected. |
| Name | Required | Description |
|---|---|---|
| id | Yes | Change request id. |
| status | Yes | The status now set on the request. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=false, destructiveHint=false, idempotentHint=true. The description adds context about the build and review flow but does not contradict annotations. No additional behavioral details beyond what annotations provide.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Single sentence, no wasted words. Front-loaded with purpose and context.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given output schema existence and simple parameters, the description sufficiently covers the tool's purpose. Could mention the status progression order but not required.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100% with descriptions for both parameters. The description does not add meaning beyond the schema; it merely restates 'update status'.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the verb 'update' and resource 'change request status' with context 'as it moves through the build and review flow'. It distinguishes from siblings like get_change_request (read) and set_preview_url (URL-specific).
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies usage for moving status along a flow, but does not explicitly state when to use or provide alternatives. No exclusions or when-not guidance.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v0.1.4get_change_request2 fields changedInput schema / properties / id / descriptionAdded value: +"Change request id, from list_change_requests."Output schema / (root)Previous value: -nullNew value: +{
+ "$schema": "http://json-schema.org/draft-07/schema#",
+ "additionalProperties": false,
+ "properties": {
+ "attachment_urls": {
+ "description": "Absolute URLs of any extra attachments.",
+ "items": {
+ "type": "string"
+ },
+ "type": "array"
+ },
+ "elements": {
+ "description": "Elements attached to the request; empty for a general request.",
+ "items": {
+ "additionalProperties": false,
+ "properties": {
+ "html": {
+ "description": "outerHTML of that element; may be empty.",
+ "type": "string"
+ },
+ "selector": {
+ "description": "CSS selector of the element the user pointed at.",
+ "type": "string"
+ }
+ },
+ "required": [
+ "selector",
+ "html"
+ ],
+ "type": "object"
+ },
+ "type": "array"
+ },
+ "id": {
+ "description": "Change request id.",
+ "type": "string"
+ },
+ "page_url": {
+ "description": "URL of the page the request was made on.",
+ "type": "string"
+ },
+ "request_text": {
+ "description": "What the user wants.",
+ "type": "string"
+ },
+ "requested_by": {
+ "description": "Requester email, or \"anonymous\".",
+ "type": "string"
+ },
+ "revision_comment": {
+ "description": "The revision note, or empty string if none.",
+ "type": "string"
+ },
+ "revision_requested": {
+ "description": "True if the requester reviewed a preview and asked for changes; push to the same branch/PR.",
+ "type": "boolean"
+ },
+ "screenshot_url": {
+ "description": "Absolute screenshot URL, or empty string.",
+ "type": "string"
+ },
+ "status": {
+ "description": "Current status.",
+ "type": "string"
+ }
+ },
+ "required": [
+ "id",
+ "status",
+ "page_url",
+ "requested_by",
+ "request_text",
+ "revision_requested",
+ "revision_comment",
+ "elements",
+ "screenshot_url",
+ "attachment_urls"
+ ],
+ "type": "object"
+}list_change_requests2 fields changedInput schema / properties / status / descriptionAdded value: +"Optional status filter: new, accepted, building, preview, approved, shipped, or rejected. Omit to list every open request."Output schema / (root)Previous value: -nullNew value: +{
+ "$schema": "http://json-schema.org/draft-07/schema#",
+ "additionalProperties": false,
+ "properties": {
+ "requests": {
+ "description": "Matching change requests, newest first.",
+ "items": {
+ "additionalProperties": false,
+ "properties": {
+ "id": {
+ "description": "Change request id; pass to get_change_request.",
+ "type": "string"
+ },
+ "page_url": {
+ "description": "URL of the page the request was made on.",
+ "type": "string"
+ },
+ "request_text": {
+ "description": "What the user asked for, in their own words.",
+ "type": "string"
+ },
+ "status": {
+ "description": "Current status of the request.",
+ "type": "string"
+ }
+ },
+ "required": [
+ "id",
+ "status",
+ "request_text",
+ "page_url"
+ ],
+ "type": "object"
+ },
+ "type": "array"
+ }
+ },
+ "required": [
+ "requests"
+ ],
+ "type": "object"
+}publish_preview2 fields changedInput schema / properties / id / descriptionAdded value: +"Change request id whose local preview is ready."Output schema / (root)Previous value: -nullNew value: +{
+ "$schema": "http://json-schema.org/draft-07/schema#",
+ "additionalProperties": false,
+ "properties": {
+ "id": {
+ "description": "Change request id.",
+ "type": "string"
+ },
+ "review_url": {
+ "description": "Page where the requester reviews and approves the change.",
+ "type": "string"
+ },
+ "status": {
+ "description": "New status, set to \"preview\".",
+ "type": "string"
+ }
+ },
+ "required": [
+ "id",
+ "status",
+ "review_url"
+ ],
+ "type": "object"
+}set_preview_url3 fields changedInput schema / properties / id / descriptionAdded value: +"Change request id."Input schema / properties / url / descriptionAdded value: +"Deploy-preview URL (from your host) that the requester will review."Output schema / (root)Previous value: -nullNew value: +{
+ "$schema": "http://json-schema.org/draft-07/schema#",
+ "additionalProperties": false,
+ "properties": {
+ "id": {
+ "description": "Change request id.",
+ "type": "string"
+ },
+ "review_url": {
+ "description": "Page where the requester reviews and approves the change.",
+ "type": "string"
+ },
+ "status": {
+ "description": "New status, set to \"preview\".",
+ "type": "string"
+ }
+ },
+ "required": [
+ "id",
+ "status",
+ "review_url"
+ ],
+ "type": "object"
+}ship_change2 fields changedInput schema / properties / id / descriptionAdded value: +"Change request id to ship (should be approved first)."Output schema / (root)Previous value: -nullNew value: +{
+ "$schema": "http://json-schema.org/draft-07/schema#",
+ "additionalProperties": false,
+ "properties": {
+ "id": {
+ "description": "Change request id.",
+ "type": "string"
+ },
+ "into": {
+ "description": "Base branch it was merged into.",
+ "type": "string"
+ },
+ "merged": {
+ "description": "Branch that was merged.",
+ "type": "string"
+ },
+ "status": {
+ "description": "New status, set to \"shipped\".",
+ "type": "string"
+ }
+ },
+ "required": [
+ "id",
+ "merged",
+ "into",
+ "status"
+ ],
+ "type": "object"
+}start_build2 fields changedInput schema / properties / id / descriptionAdded value: +"Change request id to start a local build for."Output schema / (root)Previous value: -nullNew value: +{
+ "$schema": "http://json-schema.org/draft-07/schema#",
+ "additionalProperties": false,
+ "properties": {
+ "base": {
+ "description": "Base branch it was cut from.",
+ "type": "string"
+ },
+ "branch": {
+ "description": "Branch that was created.",
+ "type": "string"
+ },
+ "id": {
+ "description": "Change request id.",
+ "type": "string"
+ },
+ "preview_url": {
+ "description": "URL that serves the worktree live for review.",
+ "type": "string"
+ },
+ "worktree": {
+ "description": "Local folder to edit; commit your change here.",
+ "type": "string"
+ }
+ },
+ "required": [
+ "id",
+ "branch",
+ "base",
+ "worktree",
+ "preview_url"
+ ],
+ "type": "object"
+}update_status3 fields changedInput schema / properties / id / descriptionAdded value: +"Change request id."Input schema / properties / status / descriptionAdded value: +"New status: new, accepted, building, preview, approved, shipped, or rejected."Output schema / (root)Previous value: -nullNew value: +{
+ "$schema": "http://json-schema.org/draft-07/schema#",
+ "additionalProperties": false,
+ "properties": {
+ "id": {
+ "description": "Change request id.",
+ "type": "string"
+ },
+ "status": {
+ "description": "The status now set on the request.",
+ "type": "string"
+ }
+ },
+ "required": [
+ "id",
+ "status"
+ ],
+ "type": "object"
+}v0.1.0get_change_requestlist_change_requestspublish_previewset_preview_urlship_changestart_buildupdate_statusScored across 7 tools
Most tools have distinct purposes, but publish_preview and set_preview_url both mark a change request as ready for review, potentially confusing an agent. start_build is a separate local-mode flow that could be misused if the normal GitHub flow is expected.
All tool names follow a consistent verb_noun pattern in snake_case (e.g., get_change_request, publish_preview, ship_change), making them predictable and easy to understand.
With 7 tools covering listing, detail retrieval, status updates, preview management, and final shipping, the count is well-scoped for managing change requests without excess or deficiency.
The tool set covers the main lifecycle but lacks a creation tool, which is acceptable if change requests are always created externally. update_status provides flexibility for status transitions.
Build and supervise fleets of agents from Claude Code, Codex or Cursor. Connects over OAuth.
Build and deploy websites, Telegram and Discord bots from chat via the DreamAgent platform.
Host files from coding agents; stage on a branch and attach to GitHub PRs.
Connect AI assistants to GitHub - manage repos, issues, PRs, and workflows through natural language.
