by duqaXxX · Agent Tool · ★ 46
seedeep See deep into what Claude Code is doing. Claude Code writes a session log to disk as it runs. tails that log and rebuilds the turn while it is still happening: the context window filling, every API call with its latency and token split, every tool call and its result, and each subagent folded under the spawn that launched it. The same view is there afterwards to walk through. Read-only. No proxy, no daemon, no session content leaves the machine.
| Stars | 46 |
| Forks | 3 |
| Language | TypeScript |
| Category | Agent Tool |
| License | MIT |
| Quality Score | 65.7668051462978/100 |
| Open Issues | 3 |
| Last Updated | 2026-09-26 |
| Created | 2026-08-07 |
| Platforms | claude-code, node |
| Est. Tokens | ~17k |
Explore other popular agent tool tools:
seedeep is What Claude Code is doing between your prompt and its answer, drawn live from the logs it already writes: every model call, every tool, each subagent on its own context window and its own model, and w. It is categorized as a Agent Tool with 46 GitHub stars.
seedeep is primarily written in TypeScript. It covers topics such as agent-observability, agentic-coding, ai-agents.
You can find installation instructions and usage details in the seedeep GitHub repository at github.com/duqaXxX/seedeep. The project has 46 stars and 3 forks, indicating an active community.
seedeep is released under the MIT license, making it free to use and modify according to the license terms.