# AI use cases for Agent Beacon (Asymptote-Labs/agent-beacon) Repo: [Asymptote-Labs/agent-beacon](https://github.com/Asymptote-Labs/agent-beacon) | Use cases: 3 | Page: https://dedupedai.com/use-cases/asymptote-labs-agent-beacon/ Beacon is an open-source, local-first memory layer for AI coding agents. It captures session history from Claude Code, Cursor, Codex, OpenCode, Cline and 20+ other harnesses into one OpenTelemetry-based trace format, then turns useful workflows, corrections and debugging patterns into reviewed knowledge that future agents retrieve through MCP or Agent Skills. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Share memory between Claude Code, Codex and Cursor: https://dedupedai.com/use-cases/asymptote-labs-agent-beacon/share-memory-between-claude-code-and-codex/ - Replay exactly what your coding agent did: https://dedupedai.com/use-cases/asymptote-labs-agent-beacon/replay-what-your-coding-agent-did/ - Send coding agent activity to your SIEM: https://dedupedai.com/use-cases/asymptote-labs-agent-beacon/forward-agent-sessions-to-your-siem/ ## How to share memory between Claude Code, Codex and Cursor with Agent Beacon Repo: [Asymptote-Labs/agent-beacon](https://github.com/Asymptote-Labs/agent-beacon) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Capture sessions from every coding agent so a fix or correction learned in one is recalled by the next. **Who it is for:** A developer who corrects the same mistake in Claude Code, then has to correct it again in Codex or Cursor. **The problem:** Each harness keeps its own memory, so a problem solved by one agent is learned from scratch by another. **The outcome:** One shared project memory that Claude Code, Codex, Cursor and OpenCode can recall through Agent Skills or MCP. ### Steps 1. Paste `Read https://beacon.sh/start.md then set me up on Beacon` into your coding agent, or on macOS run `brew tap asymptote-labs/tap` and `brew install beacon` 2. Run `beacon endpoint install` and pick Local to keep history on this machine, or confirm Beacon Cloud 3. Let setup install the Beacon Skills into ~/.agents/skills and ~/.claude/skills, or add them later with `beacon skills install` 4. Use Claude Code, Codex or Cursor normally while Beacon captures session history in the background 5. With Beacon Cloud, run `beacon mcp connect` to register the beacon-managed MCP server in Claude Code, Codex CLI, Cursor, VS Code, Gemini CLI and OpenCode; locally, `beacon mcp serve` never touches the network ### Prompt to paste into your agent ```text Set me up on Asymptote-Labs/agent-beacon using the Local destination, install the Beacon Skills, and confirm that a session in shows up in `beacon traces`. Then, in , recall what Beacon learned about and tell me which past session it came from. ``` ### Caveat Knowledge goes through a review and approve step before agents reuse it, so expect to curate what Beacon extracts rather than get perfect memory on day one. Risk, credentials: Beacon records prompts, responses, commands and file edits, so pasted API keys or secrets can end up in the trace. Interactive setup preselects Beacon Cloud, which forwards history to Asymptote Labs once you confirm; choose Local to keep everything on your machine. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/asymptote-labs-agent-beacon/share-memory-between-claude-code-and-codex/ ## How to replay exactly what your coding agent did with Agent Beacon Repo: [Asymptote-Labs/agent-beacon](https://github.com/Asymptote-Labs/agent-beacon) | Difficulty: Intermediate | Time: 20 minutes | Basis: From the README Open one trace per session with prompts, tool calls, commands, edits, approvals, MCP activity and tokens. **Who it is for:** A developer whose agent changed files overnight and who wants to see every command and approval before merging. **The problem:** Agent sessions vanish when they end, and each tool keeps its own partial, proprietary history. **The outcome:** A local, read-only replay of each session across harnesses, plus the raw JSONL to search yourself. ### Steps 1. Install Beacon and run `beacon endpoint install`, choosing Local 2. Run your agent as usual in Claude Code, Codex, Cursor or another supported harness 3. Run `beacon traces` to browse traces, event timelines, token usage and retained content in the terminal; nothing is sent anywhere 4. Run `beacon endpoint dashboard` for the local web view and open a session's lenses, such as files changed with diffs or an itemized token bill 5. Read or grep the raw event stream in ~/.beacon/endpoint/logs/runtime.jsonl ### Prompt to paste into your agent ```text Using Asymptote-Labs/agent-beacon on this machine, open the trace for my last session with `beacon traces` and list every command it ran, every file it edited and every approval, with total tokens. Verify the counts against ~/.beacon/endpoint/logs/runtime.jsonl. ``` ### Caveat Coverage differs by harness: for example, Codex CLI does not report file activity and Cursor does not report tokens in the support table. Risk, credentials: Traces retain prompt and file content, so anyone who can read ~/.beacon or install a lens can see it; the README warns to install only lenses you trust with your traces. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/asymptote-labs-agent-beacon/replay-what-your-coding-agent-did/ ## How to send coding agent activity to your SIEM with Agent Beacon Repo: [Asymptote-Labs/agent-beacon](https://github.com/Asymptote-Labs/agent-beacon) | Difficulty: Advanced | Time: 1 hour | Basis: From the README Forward normalized agent session history to Splunk, Datadog, Elastic, Sentinel or a bucket for security review. **Who it is for:** A security engineer asked which commands and MCP calls the company's coding agents ran last week. **The problem:** Every coding tool logs differently, if at all, so agent activity is invisible to the security stack. **The outcome:** One OpenTelemetry-based stream of agent sessions, tool calls, commands and approvals in the SIEM you already run. ### Steps 1. Install Beacon on each machine; system, package, MDM and CI install paths are noninteractive and account-free 2. During `beacon endpoint install` (or a repair), set up endpoint forwarding to Splunk HEC or CrowdStrike Falcon LogScale HEC 3. For Microsoft Sentinel, Sumo Logic, Wazuh or Rapid7 InsightIDR, apply the matching Beacon content pack; for Datadog or Elastic, collect the local JSONL with their agent 4. Keep the local JSONL as the default copy, or ship it to AWS S3 or Google Cloud Storage 5. Review sessions in the dashboard's security review lens, grouped by risk ### Prompt to paste into your agent ```text Install Asymptote-Labs/agent-beacon on and forward its endpoint telemetry to . Run one session, then confirm the session's prompts, commands and approvals arrive in the SIEM and match ~/.beacon/endpoint/logs/runtime.jsonl. ``` ### Caveat The README lists the supported destinations and content packs but leaves the per-SIEM steps to docs.beacon.sh, so plan to follow those pages for your destination. Risk, credentials: Forwarding sends prompts, commands and file activity off the machine, and setup needs ingest tokens such as a Splunk or LogScale HEC token; store them as secrets and tell developers what is captured. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/asymptote-labs-agent-beacon/forward-agent-sessions-to-your-siem/