# context-synthesizer

context-synthesizer synthesizes multidimensional context across task constraints, domain patterns, and past execution memory. Helps maintain clear situational awareness during complex execution.

- subagent · `type=agent` · Claude Code, Codex, Cursor · ★ 850 repo stars
- Author: ruvnet
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/ruvnet/agentic-flow/tree/HEAD/.claude/agents/reasoning/context-synthesizer.md
- Updated: 2026-10-26
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/agents/context-synthesizer

## What it does

This subagent aggregates distinct information layers: the immediate task with its constraints, domain-level practices, agent state, and historical execution records. Lifecycle hooks invoke `agentic-flow` through `npx` before and after execution to query and update the `reasoningbank` storage.

Relevance filtering uses weighted scoring across four factors: task alignment accounts for 40%, reliability takes 25%, actionability gets 20%, and recency contributes 15%. Items below the 0.6 threshold get pruned, keeping at most 20 entries per category to avoid context bloat.

Without Node.js and an accessible `npx` binary in the environment, the pre- and post-execution hooks will fail during memory retrieval.

## Install

### Claude Code (native platform)

**Manually: put the files in place**
- File: `.claude/agents/context-synthesizer.md` — For all projects use ~/.claude/agents/. Claude delegates by the description field or an @-mention

### Codex

**Manually: put the files in place**
- File: `.codex/agents/context-synthesizer.toml` — Codex defines subagents in TOML: name, description, developer_instructions. Move the Markdown text into developer_instructions by hand. For all projects use ~/.codex/agents/

### Cursor

**Manually: put the files in place**
- File: `.cursor/agents/context-synthesizer.md` — Cursor also reads .claude/agents/

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/ruvnet/agentic-flow/tree/HEAD/.claude/agents/reasoning/context-synthesizer.md

## When to use it
- Consolidate fragmented task requirements and domain knowledge into one brief
- Filter out irrelevant historical memories using weighted scoring criteria
- Retrieve past failure modes and patterns from `reasoningbank`
- Prepare structured risk warnings and execution strategies for peer agents

## Example prompts
- Synthesize context for refactoring the authentication layer using past project memories.
- Gather domain best practices and environmental constraints before tackling this migration.
- Analyze previous task outcomes and build an execution summary for this feature.

## Similar
- [context-engineering-collection](https://affpapa.org/en/nevibecode/skills/context-engineering-collection.md) — skill · 4500 installs
- [context-manager](https://affpapa.org/en/nevibecode/agents/voltagent-context-manager.md) — subagent · ★ 6505 repo stars
- [agent-startup-and-context-ingestion](https://affpapa.org/en/nevibecode/skills/agent-startup-and-context-ingestion.md) — skill · ★ 8928 repo stars
- [Swarm Orchestration](https://affpapa.org/en/nevibecode/skills/swarm-orchestration.md) — skill · 2100 installs
- [knowledge-synthesizer](https://affpapa.org/en/nevibecode/agents/voltagent-knowledge-synthesizer.md) — subagent · ★ 6505 repo stars
- [llm-context](https://affpapa.org/en/nevibecode/skills/llm-context.md) — skill · 1300 installs

From the same set: [ruvnet/agentic-flow](https://affpapa.org/en/nevibecode/by/ruvnet/agentic-flow.md) — 76 cards

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
