Canonical: https://botbento.com/usecases/coding-agent-context-retriever-with-code-aware/
Format: Markdown representation of the public HTML page.

# Coding agent context retriever with code-aware memory

[Coding](https://botbento.com/usecases/coding/) · Intermediate

Finds relevant code and past decisions for your coding agent.

An AI bot indexes a codebase and stores project memories. It answers a coding agent's queries with the most relevant code snippets and context, including file and line references. It ties memories to code locations so outdated information can be flagged when code changes.

## Quick answers

**What does it do?** Finds relevant code and past decisions for your coding agent. Result: Return snippets with file and line.

**What does it need?** Tools: GitHub, Terminal, Docs, Database. Intermediate: some setup. The prompt includes an approval step.

**How do I run it?** When something happens: Agent asks for context. Copy the prompt below into a bot in BotBento.

## How it works

1. When something happens: Agent asks for context
2. Then: Search code map and memory
3. Result: Return snippets with file and line

## Prompt to start with

```text
You are a context retrieval bot for a coding agent. Index the repository's code structure and store project decisions and past session notes in a memory store. When the coding agent asks for help, search both the code map and memory store, and return the most relevant code snippets with file and line references. If a memory is tied to code that has changed, flag it as possibly outdated. Do not send messages, make payments, or delete any data without the user's explicit approval.
```

## Tools it uses

GitHub, Terminal, Docs, Database

## Works with

BotBento

## Tags

coding, memory, code-search, context, developer-tools

## Source

[Show HN: Graphicle, a Rust port of Graphify with memory for coding agents](https://news.ycombinator.com/item?id=50088713) — Hacker News · checked 2026-11-13

Summarized in our own words. Any results described there are the authors’ own, not measured by BotBento.

Verified 2026-11-13: passed every automated gate and a second-model review (kimi-code/k3.1-pro). Automated, not human review. [How verification works](https://botbento.com/usecases/verification/)

By BotBento Editorial · Added 2026-11-13 · Updated 2026-11-13 · [Suggest a correction](https://botbento.com/contact/)

## Related use cases

- [Persistent local memory for coding sessions](https://botbento.com/usecases/persistent-local-memory-for-coding-sessions/) — Track repository architecture, conventions, and past decisions across coding sessions
- [Extract codebase rules from developer session logs](https://botbento.com/usecases/extract-codebase-rules-from-developer-session-logs/) — Turn recurring coding errors into path-scoped repository guidelines
- [Summarize project architecture before coding](https://botbento.com/usecases/initializing-coding-sessions-with-system-architecture-contex/) — Pull service boundaries and dependencies into a short context brief before you code
