Skip to main content
Glama
tksfjt1024

codex-cli-mcp-slim

by tksfjt1024
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Local

codex-reply

Continue a Codex session by thread ID with a follow-up prompt, returning the new final agent message. Use it to resume prior work or ask further questions.

Instructions

Continue a previous Codex session (codex exec resume <THREAD_ID>) with a follow-up prompt and return the new final message. Only the flags codex exec resume accepts are exposed; the working directory and sandbox come from the current configuration (this server's own flags and config.toml), not from the original session.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
envNoExtra environment variables for the codex subprocess.
modelNoPass -m <MODEL>: the model slug for this call, overriding config.toml.
configNoConfiguration overrides as key=value strings, one per element. Each entry maps to a separate codex -c flag (for example "model_reasoning_effort=high"). Values are parsed as TOML by codex, so quote strings that are not bare words.
promptYesPrompt sent verbatim to codex on stdin. codex runs its full agentic loop and this server returns the final agent message.
ephemeralNoPass --ephemeral: do not persist the session's rollout file to disk.
thread_idYesThe thread to continue: the thread_id from a previous result's [codex] line. Maps to codex exec resume <THREAD_ID>.
extra_argsNoRaw CLI flags appended verbatim (one token per element). Use to reach new or uncommon codex exec flags without updating this server. Do not pass --json or a prompt: the server adds both.
timeout_secondsNoHard wall-clock timeout for the codex subprocess in seconds (default 1800).
skip_git_repo_checkNoPass --skip-git-repo-check: allow running in a directory that is not inside a git repository. Without it codex refuses such a working directory.

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.1.0

TDQS

A4.2/5.0
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description carries the full burden and does meaningful work: it discloses that working directory and sandbox come from the current server configuration rather than the original session, and that it returns the new final message. It omits auth requirements and other runtime caveats, but the config-inheritance note is exactly the kind of non-obvious behavior an agent needs.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Two tight sentences, behavior-first, with the CLI mapping inline and the configuration caveat front-loaded second. No redundant restatement of parameters or boilerplate.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a 9-parameter subprocess-invoking tool with no annotations and no output schema, the description covers the key semantics (session continuation, config inheritance, return value). Remaining gaps around error/exit behavior are minor given how fully the schema describes each flag.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

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 all nine parameters thoroughly. The description reiterates the thread_id mapping and prompt behavior but adds no syntax or format detail beyond the schema, which is the expected baseline when the schema does the heavy lifting.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb (continue) and resource (a previous Codex session) and maps directly to the underlying `codex exec resume <THREAD_ID>` command. The scoping to continuation clearly separates it from the `codex` sibling that starts a fresh run.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description makes clear this resumes an existing session and requires a thread_id, implying the alternative (the `codex` sibling) is for new sessions. It stops short of explicitly naming the sibling or stating when not to use this tool, so it lands at a strong 4 rather than a 5.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Deploy Server

Other Tools