nvxtien/context-slice

> Less context. Unchanged signal.
>
> Give your coding assistant the smallest useful slice of your codebase.
When you use AI to write code, the model doesn’t spend tokens just writing the
final answer. A large part of the cost often comes from the code it has to
read first: the target file, its callers and callees, dependencies, tests, and
configuration.
Reading more code means more tokens, slower responses, and more noise around
the detail that actually matters. The practical way to reduce that cost is to
reduce how much code the agent has to read in the first place.
ContextSlice does exactly that. It builds a small, task-specific context from
Java, TypeScript, TSX, JavaScript, Python, Rust, Go, C, C++, C#, and Kotlin
source. It finds the target symbol, follows relevant callers and callees,
and reports what was included or left out.
In short: instead of giving an AI assistant the whole repository, give it the
smallest useful slice of the codebase.
It is local, read-only, and deterministic: Tree-sitter performs the structural
analysis, SQLite stores the index, and no source code is sent to a hosted
service by ContextSlice.
The result is:
- Less code to read and pay for: return the target symbol plus the most relevant callers and callees instead of whole files.
- Better signal: rank results by task relevance and keep unresolved runtime dispatch explicit instead of guessing.
- Predictable output: enforce a token budget and explain why items were included or omitted.
- Local control: keep source analysis and the SQLite index in the repository; the target source is never edited.
Quick start
Claude Code plugin
Install it directly from the GitHub marketplace:
The plugin provides a skill that tells Claude Code to request a focused
ContextSlice preview before reading source files. Its PreToolUse hook also
blocks supported source reads until context.preview has completed, so this
is enforced at runtime rather than relying only on prompt compliance. The MCP
server then targets the project Claude Code has open. The plugin starts the published
the latest published context-slice npm runtime, so Claude needs npm registry access on first
use. The package installs native dependencies for the current platform; later
runs use the local npm cache.
Codex plugin
Add the GitHub marketplace to Codex:
Then open /plugins, choose Context Slice Marketplace, and install
context-slice. The plugin includes the same ContextSlice skill and stdio MCP
server for the project Codex has open.
How it works
1. Discover a repository and scan supported source while ignoring common generated/build directories.
2. Store symbols, call edges, hashes, schema version, and refresh time in a local SQLite cache.
3. Refresh before preview or MCP tool execution so changed source files are not silently served stale.
4. Select a target from task text, then include the target body plus ranked direct callers/callees until the strict token budget is full.
Available MCP tools are context.search, context.symbol, context.callers, context.preview, context.slice, and context.diff. MCP stdout contains protocol messages only; diagnostics must not corrupt stdio framing.
CLI
| Command | Purpose |
| ------------------- | ----------------------------------------------------------- |
| init | Discover the repository and create/refresh the local index. |
| index | Refresh the index explicitly. |
| status | Show readiness, schema, cache freshness, and last refresh. |
| doctor | Check repository, source, cache, and MCP command readiness. |
| preview | Return a deterministic, strict-budget context preview. |
| mark-dirty [file] | Tell a running MCP index that source changed. |
| mcp | Start the stdio MCP server with the stable public command. |