This server lets you run multiple Codex CLI tasks in parallel and monitor/manage them from any MCP client, with drop-in compatibility for the official Codex MCP tool.
Synchronously run a Codex session (codex)
Launch background tasks that run in parallel and return a task_id (codex_async)
Wait for one or more tasks to finish and collect all results in one call (codex_wait)
Continue previous sessions with a follow-up prompt using threadId (codex_reply)
See live progress of running tasks: phase, elapsed time, tools called, last command, current thinking (codex_status)
Cancel a running task while preserving its worktree and partial output (codex_cancel)
Isolate each task in its own git worktree to prevent parallel tasks from conflicting
Override per-call settings: model, sandbox mode, approval policy, working directory, profile, config, instructions
Set server-level defaults via -c key=value flags, mirroring codex mcp-server
Use MCP resources to inspect server info, config, and task state
Works without pip dependencies (stdlib only) and without a local git repo (--skip-git-repo-check)
Wraps OpenAI's Codex CLI for parallel execution, live monitoring, and worktree isolation.
Click on "Install 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., "@codex-mcp-swarm run three parallel code reviews with worktree isolation"
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.
An MCP server that wraps OpenAI's Codex CLI with true parallel execution and live task monitoring. Zero dependencies -- single Python file, stdlib only.
The official codex mcp-server processes requests sequentially. If your MCP client (Claude Code, etc.) needs to run 5 Codex tasks, they queue up one after another. This server spawns each task as an independent subprocess, so they run in parallel.
Unique features no other Codex MCP wrapper has:
Worktree isolation -- worktree: true creates an isolated git worktree per task so parallel Codex instances never edit past each other
Batch wait -- launch N tasks, call codex_wait once, get all results when they finish
Live status -- see what each Codex task is doing right now (last tool call, current reasoning, progress)
Full flag parity -- same parameters as the official Codex MCP tool (sandbox, approval-policy, cwd, model, config, etc.)
Drop-in config -- accepts the same -c key=value server args as codex mcp-server
Related MCP server: switchyard
Tool | Description |
| Synchronous execution (drop-in replacement for official); terminal failures include the model, structured cause, category, and retryability |
| Launch a task, get a |
| Continue a previous session via |
| Live view: tools called, last command, current thinking, warnings, and terminal failure cause |
| Block until multiple tasks complete and return results or structured failure details |
| Kill a running async task (preserves worktree for inspection); if it already failed, return the recorded cause |
claude mcp add codex-swarm -- uvx --upgrade codex-mcp-swarm \
-c model=gpt-6-astra \
-c approval_policy=never \
-c sandbox_mode=danger-full-access \
--skip-git-repo-checkThat's it. No clone, no setup. uvx downloads and runs it directly from PyPI. The --upgrade flag ensures you always get the latest version on restart.
Note: Requires uv (
curl -LsSf https://astral.sh/uv/install.sh | sh). Alternatively, usepipx run codex-mcp-swarminstead ofuvx codex-mcp-swarm.
~/.claude.json){
"mcpServers": {
"codex-swarm": {
"type": "stdio",
"command": "uvx",
"args": [
"--upgrade",
"codex-mcp-swarm",
"-c", "model=gpt-6-astra",
"-c", "approval_policy=never",
"-c", "sandbox_mode=danger-full-access",
"--skip-git-repo-check"
]
}
}
}The -c flags are identical to codex mcp-server -- copy-paste your existing config.
1. Call codex_async with prompt A --> task_id: "abc123"
2. Call codex_async with prompt B --> task_id: "def456"
3. Call codex_async with prompt C --> task_id: "ghi789"
4. Call codex_wait(task_ids=["abc123", "def456", "ghi789"])
--> blocks until all finish, returns all resultsA task_id is internal to this server. Your MCP client does not track it.
It will not appear in a task list, and no notification fires when the task
finishes. If the agent's turn ends before codex_wait is called, the Codex run
still completes and writes its result to disk, but nothing is left to deliver
it and nothing wakes the session. codex_status is a progress peek only; it
collects nothing.
So: codex_async and codex_wait belong in the same turn.
Let the wait run long. Claude Code moves any main-conversation tool call still
running after two minutes into a tracked background task and re-invokes the
session with the result when it settles -- that is the wake-up mechanism, so
crossing the two-minute line is the goal rather than something to dodge. Short
timeout values are raised to a floor (default 150s, override with
CODEX_SWARM_MIN_WAIT) for exactly this reason. Tasks that have already
finished return instantly regardless of the floor.
codex exec writes progress, warnings and sandbox notices to stderr on
perfectly healthy runs. Until 1.10.0, a task whose exit code was lost to a
reaping race was marked failed on the strength of non-empty stderr alone,
so finished work got reported as a failure and callers acted on it.
Version 1.11.0 uses the terminal events in codex exec --json as the primary
lifecycle signal. An unambiguous turn.completed means success even if an
earlier top-level error event was emitted while the CLI retried. An
unambiguous turn.failed means failure even if an earlier, partial
agent_message exists. Top-level error events are diagnostic rather than
terminal, because the public JSONL omits the CLI's internal will_retry field;
item.completed events whose item type is error are warnings. If terminal
events are absent or contradict each other, an observed process exit code
decides the result. Only a verdict made without either signal is marked as
inferred, and both completion and failure outputs label that explicitly.
Terminal failures report the model parsed from the command that actually ran,
the structured turn.failed.error.message, a failure category, whether a later
retry is likely to be useful, and a suggested action. Retryability is
information only: the wrapper never starts a second Codex turn. The Codex CLI
may perform its own in-turn retries. Stderr can be included as a clearly
labeled, truncated diagnostics section, but it never replaces the primary
failure cause.
Two things follow for anyone consuming this server:
A FAILED from codex_status or codex_wait that is flagged as inferred is
a prompt to verify, not a conclusion. Check pgrep -af 'codex exec', file
mtimes, the worktree branch, and your own build or tests.
A codex_wait that times out is not a failure at all. The task is not
killed. Call codex_wait again with the same task_id.
Prevent parallel tasks from editing the same files:
1. Call codex_async(prompt="Refactor auth", worktree=true)
--> task_id: "abc123"
--> Worktree Branch: codex-swarm/abc123
2. Call codex_async(prompt="Add logging", worktree=true)
--> task_id: "def456"
--> Worktree Branch: codex-swarm/def456
3. codex_wait(task_ids=["abc123", "def456"])
4. git merge codex-swarm/abc123
5. git merge codex-swarm/def456Each task gets its own git worktree and branch based on HEAD. After completion, merge the branches back. Worktrees are automatically cleaned up after 24 hours (configurable via CODEX_SWARM_TASK_MAX_AGE).
Call codex_status(task_ids=["abc123"])
-->
=== Task abc123 (45s elapsed) ===
Phase: running
Tools called: 23
Last tool: exec_command(grep -rn "handleError" src/)
Output: Analyzing error handling patterns across the codebase...1. Call codex(prompt="Review this file") --> result + session persisted
2. Call codex_reply(threadId="<session-uuid>", prompt="Now fix the bug you found")For a follow-up that will run more than a few minutes, resume in the
background instead -- codex_async accepts threadId and gives you a
task_id that survives a client idle timeout:
1. Call codex_async(threadId="<session-uuid>", prompt="Now implement it")
--> task_id: "abc123"
2. Call codex_wait(task_ids=["abc123"]) --> result, with full prior contextcodex_reply is synchronous and emits progress notifications while it waits,
but a client that gives up on silence will still abandon the request. Only the
async path leaves you a handle to recover with.
Flag | Description |
| Config default (repeatable). Same format as |
| Allow running outside git repos. |
| Don't persist session files. Disables |
All parameters from the official Codex MCP tool are supported:
prompt (required)
model -- override server default
sandbox -- read-only, workspace-write, danger-full-access
approval-policy -- untrusted, on-failure, on-request, never
cwd -- working directory
profile -- config profile from config.toml (ignored on threadId resumes -- codex exec resume has no --profile flag)
config -- object of key=value overrides
worktree -- run in an isolated git worktree (prevents parallel tasks from conflicting)
base-instructions, developer-instructions, compact-prompt
The server exposes read-only resources for discoverability:
URI | Description |
| Version, capabilities, directories, config |
| Current server-level defaults and flags |
| All known tasks and their current state; failed entries include model, cause, category, and retryability |
Variable | Default | Description |
|
| Log file path |
|
| Log level ( |
|
| Task output storage directory |
|
| Worktree storage directory |
|
| Seconds before completed task artifacts (and worktrees) are cleaned up |
|
| Floor in seconds for |
Python 3.8+
Codex CLI installed and authenticated
No pip dependencies (stdlib only)
Works on Linux and macOS (Linux gets extra PID reuse protection and zombie detection via /proc)
Originally inspired by jeanchristophe13v/codex-mcp-async. Rewritten with full flag parity, JSONL status parsing, batch wait, and session reply support.
MIT
Run a Codex session synchronously. Parameters match the official Codex MCP tool.
| Name | Required | Description | Default |
|---|---|---|---|
| prompt | Yes | The initial user prompt for the Codex session. | |
| approval-policy | No | Approval policy for shell commands generated by the model. | |
| sandbox | No | Sandbox mode. | |
| cwd | No | Working directory for the session. If relative, resolved against the server's cwd. | |
| worktree | No | Create an isolated git worktree and branch for this task. Each task gets its own copy of the repo so parallel tasks never conflict. The response includes the branch name (codex-swarm/<task_id>) -- merge it back when done. | |
| model | No | Optional override for the model name (e.g. 'gpt-5.4'). | |
| profile | No | Configuration profile from config.toml. | |
| config | No | Config settings that override server defaults. | |
| base-instructions | No | Instructions to use instead of the defaults. | |
| developer-instructions | No | Developer instructions injected as developer role message. | |
| compact-prompt | No | Prompt used when compacting the conversation. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries full burden for behavioral disclosure. It only says 'synchronously', omitting details like blocking, duration, side effects (sandbox, worktree creation), or return value. This is insufficient for safe invocation.
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, front-loaded sentence that efficiently states the core purpose. It is appropriately concise, though could be slightly more specific.
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 11 parameters, no output schema, and no annotations, the description is too minimal. It fails to explain what a Codex session does, what it returns, or important behaviors (e.g., it may take time, creates worktrees).
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 baseline is 3. The description adds no additional meaning beyond what the schema already provides. The phrase 'Parameters match the official Codex MCP tool' is generic and not helpful.
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 it runs a Codex session synchronously, which is a specific verb+resource. It implicitly differentiates from siblings like codex_async by mentioning 'synchronously', but does not explicitly name alternatives.
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 provides no explicit when-to-use guidance. It implies synchronous execution, but does not contrast with async or other management tools. Agents are left to infer usage from the tool name.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Start a Codex task in the background and return immediately with a task_id. Use codex_wait to collect results from one or more tasks, or codex_status to monitor progress. Set worktree=true to isolate each task in its own git worktree so parallel tasks don't conflict -- merge the branch back when done.
| Name | Required | Description | Default |
|---|---|---|---|
| prompt | Yes | The initial user prompt for the Codex session. | |
| approval-policy | No | Approval policy for shell commands generated by the model. | |
| sandbox | No | Sandbox mode. | |
| cwd | No | Working directory for the session. If relative, resolved against the server's cwd. | |
| worktree | No | Create an isolated git worktree and branch for this task. Each task gets its own copy of the repo so parallel tasks never conflict. The response includes the branch name (codex-swarm/<task_id>) -- merge it back when done. | |
| model | No | Optional override for the model name (e.g. 'gpt-5.4'). | |
| profile | No | Configuration profile from config.toml. | |
| config | No | Config settings that override server defaults. | |
| base-instructions | No | Instructions to use instead of the defaults. | |
| developer-instructions | No | Developer instructions injected as developer role message. | |
| compact-prompt | No | Prompt used when compacting the conversation. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations exist, so the description carries full burden. It transparently explains the async behavior, immediate return, and worktree isolation. However, it does not disclose potential side effects, authorization needs, or task persistence, which are minor gaps.
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 extremely concise with three sentences, each adding distinct value. The main action is front-loaded, and every sentence earns its place without unnecessary 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 the tool's complexity (11 parameters, async workflow), the description covers the key workflow: start, get task_id, use companion tools. It lacks details on return format and error conditions, but is otherwise complete for a background task 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 coverage is 100%, so the baseline is 3. The description adds value by explaining the worktree parameter's behavior and purpose, but does not enhance understanding of other parameters beyond their schema descriptions.
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 starts a Codex task in the background and returns a task_id immediately. It distinguishes itself from siblings by referencing codex_wait and codex_status for collecting results and monitoring progress.
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?
Explicitly guides when to use: for background task execution. Mentions alternatives (codex_wait, codex_status) and provides context for using the worktree parameter to avoid conflicts in parallel tasks.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Kill a running async Codex task. The process is terminated and the task is marked as cancelled. Any worktree and partial output are preserved for inspection.
| Name | Required | Description | Default |
|---|---|---|---|
| task_id | Yes | The task_id to cancel. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations provided; description carries full burden. It discloses that the process is terminated, task marked cancelled, and worktree/partial output preserved. This is good but lacks mentions of permissions or irreversible effects.
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, front-loaded with the action. Every word serves a purpose; no wasted text.
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 low complexity (1 parameter, no output schema), the description provides sufficient behavioral and side-effect information to understand the tool's function.
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 description adds minimal value beyond the schema description. It basically restates the parameter's purpose without additional details on format or constraints.
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 (kill/cancel) and resource (async Codex task). It distinguishes from siblings like codex_async (starts tasks) and codex_status (checks status).
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?
Description implies usage for canceling a running task but provides no explicit guidance on when to use versus alternatives or when not to use.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Continue a Codex conversation by providing the thread/session ID and a follow-up prompt. Uses codex exec resume under the hood.
| Name | Required | Description | Default |
|---|---|---|---|
| threadId | Yes | The session/thread ID (UUID) from a previous Codex call. | |
| prompt | Yes | The follow-up prompt to continue the conversation. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description bears full responsibility. It mentions using 'codex exec resume' under the hood, but does not disclose whether the tool is blocking, error handling, or rate limits. This is minimal behavioral disclosure for a continuation tool.
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 plus a brief technical note. It is front-loaded with the core purpose, no wasted words, and efficient for agent consumption.
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 (2 required params, no output schema), the description is minimally viable. It explains the action and underlying command but lacks usage guidelines and behavioral details. Could be more complete with usage context.
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%, so the schema already documents both parameters thoroughly. The description adds no new meaning beyond restating the parameter roles. Baseline 3 is 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?
The description clearly states the purpose: 'Continue a Codex conversation' by providing a thread ID and prompt. It distinguishes from sibling tools like 'codex' (which likely starts new conversations) and async variants, making it easy for an agent to select appropriately.
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 use when continuing an existing conversation, and the sibling list provides context. However, it lacks explicit 'when to use' vs. alternatives, such as preferring 'codex' for new sessions. Still, the guidance is clear for most scenarios.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Get live status of running async Codex tasks. Shows what each task is currently doing: last tool call, reasoning, progress. Works on both running and completed tasks.
| Name | Required | Description | Default |
|---|---|---|---|
| task_ids | Yes | List of task_ids to check status for. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries full burden. It discloses the tool is for live status (no side effects) and specifies what information it provides. It could mention polling behavior or rate limits, but the transparency is adequate for a read-only status check.
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 sentences long, front-loaded with the main action, and contains no fluff. Every sentence provides useful information.
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?
The tool is simple (1 param, no output schema). The description covers functionality and output details. It omits the exact output format, but given no output schema, this is acceptable. Overall complete for the tool's complexity.
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 a clear description for task_ids. The description does not add extra meaning beyond the schema, but baseline 3 is appropriate since the schema already documents the parameter adequately.
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 retrieves live status of async Codex tasks, specifying what information it shows (last tool call, reasoning, progress). It differentiates from siblings like codex_async (starts tasks) and codex_cancel (cancels them).
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 after starting an async task with codex_async and notes it works on both running and completed tasks. It does not explicitly exclude use cases or state alternatives, but context from sibling names provides enough guidance.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Block until one or more async Codex tasks complete, then return all results. Accepts a list of task_ids. This avoids repeated polling -- call once after launching codex_async tasks. Default timeout is 1800s (30 min). If a task times out, it is NOT killed -- it keeps running. You can call codex_wait again with the same task_ids to resume waiting, or use codex_status to check progress.
| Name | Required | Description | Default |
|---|---|---|---|
| task_ids | Yes | List of task_ids to wait for. | |
| timeout | No | Max seconds to wait (default: 1800). The task keeps running even if this times out. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description bears full responsibility for behavioral disclosure. It clearly states the tool blocks, has a default timeout of 1800s, and that timed-out tasks continue running. This covers the key behaviors without omitting critical details.
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 extremely concise: two sentences plus a timeout note. It immediately states the purpose and provides necessary usage context without superfluous 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 the absence of an output schema and the tool's simplicity (two parameters), the description covers all essential aspects: purpose, when to use, timeout behavior, and follow-up options. It is complete for effective tool selection and invocation.
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 for parameters is 100%, with descriptions already present. The tool description adds context about default timeout and non-killing behavior, but this is also captured in the schema's timeout description. Thus, parameter semantics are adequate but not significantly enhanced beyond the schema.
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 'Block until one or more async Codex tasks complete, then return all results.' It distinguishes from siblings like codex_status and codex_async by noting that it avoids polling and contrasts with launching tasks.
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 to 'call once after launching codex_async tasks' and explains timeout behavior: 'If a task times out, it is NOT killed -- it keeps running. You can call codex_wait again... or use codex_status to check progress.' This provides clear guidance on when and how to use the tool.
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. Dates show when Glama detected each change.
v1.7.0codexcodex_asynccodex_cancelcodex_replycodex_statuscodex_waitEach tool has a clearly distinct purpose: sync execution, async launch, cancel, reply, status polling, and waiting. No two tools overlap in functionality.
Most tools follow the 'codex_' prefix with a clear verb, but the sync tool is simply 'codex' without an underscore, creating a minor inconsistency.
6 tools is appropriate for managing Codex sessions, covering sync and async operations, cancellation, continuation, status, and blocking wait.
Core lifecycle is well-covered (start, wait, cancel, continue, monitor). Minor gap: no tool to list all active tasks, though status can query known task IDs.
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
A paid remote MCP for OpenAI Codex agent coordination MCP, built to return verdicts, receipts, usage
Source-checked CLI guides and model-aware planning for Claude Code, Codex, and Grok Build.
Task tracking built for coding agents. Work is leased, so two agents never take the same SubTask.
Shared task layer for AI coding agents. One MCP surface: task_search, task_get, task_mutate.
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/TKasperczyk/codex-mcp-swarm'
If you have feedback or need assistance with the MCP directory API, please join our Discord server
