get-linked-context

Reads conversation transcripts, summaries, and terminal output from linked agent nodes on the nodeterm canvas.

1|Updated Aug 17, 2026
One-click install
npx skills add https://github.com/Marshal-Nguyen/Skill_Claude_Agent --skill get-linked-context-marshal-nguyen
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: get-linked-context
Source: https://github.com/Marshal-Nguyen/Skill_Claude_Agent/tree/main/claude/skills/get-linked-context
Command: npx skills add https://github.com/Marshal-Nguyen/Skill_Claude_Agent --skill get-linked-context-marshal-nguyen

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? When multiple AI agent sessions (Claude, Codex, Gemini) run as connected nodes on the nodeterm canvas, each session is blind to what the others have done. This Skill lets an agent pull the context of a linked node on demand, enabling handoffs, continuation of work, and coordination without manual copy-pasting. ## Core Features & Use Cases - List Linked Nodes: Run the list command to see every agent node and sticky note connected to the current session via a context-link edge. - Read Summaries and Transcripts: Fetch the last N lines of a linked node's conversation with summary, or its full transcript with transcript. - Inspect Terminal Output: Use terminal to view a linked node's recent visible terminal buffer. - Read Live Sticky Notes: Linked sticky notes appear in list marked (note) and are read live from the canvas, so their current text is always returned. - Use Case: A Claude node finishes scaffolding an API and a linked Codex node needs to continue the implementation; the Codex session runs summary to see what Claude already did before writing code. ## Quick Start Ask the agent to list the nodes it is linked to on the nodeterm canvas and summarize what the connected node has been doing.

Frequently Asked Questions about get-linked-context

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I read another agent's conversation in a nodeterm session?▼

Run the context.sh shim with the `summary` command to get the last N lines of a linked node's conversation, or `transcript` for the full conversation. Start with `list` to see which nodes you are linked to.

How do I see terminal output from a linked Claude or Codex node?▼

Use the `terminal` command of the context.sh shim, optionally passing `--node <id|title>` to target a specific linked node. It returns the node's recent visible terminal buffer.

Does get-linked-context work outside a nodeterm session?▼

No. The Skill is only meaningful inside a nodeterm session with a context-link edge. If the CLI reports "Not a nodeterm session" or "No linked nodes", there is nothing to read and you should not retry.

When do I need to pass the --node flag to context.sh?▼

The `--node` flag is optional when you are linked to exactly one node. When linked to multiple nodes, pass the id or title shown by the `list` command to select the target.

Can I read sticky notes linked to my agent node?▼

Yes. Linked sticky notes appear in `list` marked `(note)`, and running `summary` or `transcript` on a note prints its current text, read live from the canvas.