
Sequential Thinking
- 21 repo stars
- Updated August 5, 2026
- dashed/claude-marketplace
Give an agent a single sequentialthinking tool for dynamic, reflective, step-by-step reasoning with branching and revision of earlier steps.
About
This MCP server exposes a single sequentialthinking tool for dynamic, reflective, step-by-step problem-solving. Use it when a task needs structured reasoning, planning, hypothesis generation, branching to explore alternatives, or revising earlier steps while keeping a running chain of thoughts. It augments an agent's reasoning process rather than performing a domain task.
- MCP sequentialthinking tool
- Structured step-by-step reasoning
- Branching to explore alternatives
- Revise earlier steps, running chain
Sequential Thinking by the numbers
- Data as of Aug 5, 2026 (Skillselion catalog sync)
/plugin marketplace add dashed/claude-marketplace/plugin install sequential-thinking@alberto-marketplaceAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| repo stars | ★ 21 |
|---|---|
| Last updated | August 5, 2026 |
| Repository | dashed/claude-marketplace ↗ |
What it does
Give an agent a single sequentialthinking tool for dynamic, reflective, step-by-step reasoning with branching and revision of earlier steps.
README.md
sequential-thinking
An MCP server that exposes a single
sequentialthinking tool for dynamic, reflective, step-by-step
problem-solving. It is a Python port of the official TypeScript
sequential-thinking MCP server.
This is the first MCP-server plugin in this marketplace — it ships a
runnable server instead of a SKILL.md.
What it does
The server registers one tool with the model:
sequentialthinking— Record one step in an iterative thinking process. The model can record an evolving chain of thoughts, branch off to explore alternatives, revise earlier steps, and adjust the planned thought count on the fly. History is kept in memory for the lifetime of the process.
Once the plugin is enabled, Claude Code namespaces the tool as
mcp__plugin_sequential-thinking_sequential-thinking__sequentialthinking
(pattern: mcp__plugin_<plugin>_<server>__<tool>). Only a directly-configured
(non-plugin) MCP server registers the bare
mcp__sequential-thinking__sequentialthinking form.
Tool parameters
| Parameter | Type | Description |
|---|---|---|
thought |
string | Your current thinking step |
nextThoughtNeeded |
bool | Whether another thought step is needed |
thoughtNumber |
int | Current thought number (e.g., 1, 2, 3) |
totalThoughts |
int | Estimated total thoughts needed |
isRevision |
bool (optional) | Whether this revises previous thinking |
revisesThought |
int (optional) | Which thought is being reconsidered |
branchFromThought |
int (optional) | Branching point thought number |
branchId |
string (optional) | Branch identifier |
needsMoreThoughts |
bool (optional) | If more thoughts are needed |
Numeric and boolean arguments are also accepted as JSON strings (e.g. "1",
"true"), working around clients — including Claude Code — that serialize
these values as strings.
Environment variables
DISABLE_THOUGHT_LOGGING— When set (case-insensitive) to"true", the server suppresses the formatted thought box that is otherwise written to stderr. All other behavior is unchanged.For a plugin install, set it by adding an
envblock to the server entry in the plugin's.mcp.json:"env": { "DISABLE_THOUGHT_LOGGING": "true" }
Prerequisites
- uv — the server is launched with
uv run --script. The script is a PEP 723 inline-dependency script (requires-python >=3.10,mcp>=1.9.4,<2); uv resolves and caches the dependencies automatically on first run. No manualpip installis needed.
How it is wired
The plugin declares the server in .mcp.json:
{
"mcpServers": {
"sequential-thinking": {
"command": "uv",
"args": ["run", "--no-config", "--script", "${CLAUDE_PLUGIN_ROOT}/scripts/mcp_sequential_thinking.py"]
}
}
}
${CLAUDE_PLUGIN_ROOT} is expanded by Claude Code to the installed plugin
directory, so the server runs regardless of where the plugin is checked out.
--no-config keeps uv from adopting the consuming repo's
pyproject.toml/uv.toml (e.g. a private default package index), which would
otherwise break dependency resolution when Claude Code is launched from such a
directory.