Build your library
Menu

Marketplace / Agents

Code Oracle

Code Oracle answers one question about where code is, what calls it, how it flows or where to change it, from the workspace's code index.

  • Read-only
  • Models via ContextStream
  • Searches code
  • Reads memory
  • Reads context

What it does

Answers where-is, what-calls, how-does-it-flow and where-to-change questions about indexed code: a cited answer of at most 600 tokens and a read list.

It does not do these. A task like them is declined, and a declined run is free:

  • Changing code, writing patches or tests, or opening pull requests.
  • Reviewing a diff or a pull request (PR Reviewer does that).
  • Writing an onboarding guide to a whole repository (Repo Guide does that).
  • Plan and task status: what is next, what is blocked, who is on what (Plans & Tasks Keeper does that).
  • Returning file contents beyond short quotes; it returns a read list instead.
  • Questions that are not about the indexed code.

Real runs

What Code Oracle did with a real input: the input, the result (shortened where it is long), how long it took, what it cost, and where it ran. Each live run used public or demo material only. An example marked as from the eval suite is not a live run.

Live run on ContextStream’s task rail

What calls `index_status_from_data`?

Agent
oracle@0.2.0
Outcome
succeeded
Ran
2026-10-27 17:18 UTC
Took
3.9 s
Cost
Free

ContextStream's Demos workspace (demo material only), in a project that indexes the public repository github.com/contextstream/mcp-server at v1.0.34.

Input
{
  "project_id": "6aad1065-8086-4114-931e-1b16c1d9b234",
  "question": "What calls `index_status_from_data`?"
}

Result

Answer

`index_status_from_data` (defined at crates/mcp-server/src/hook_handlers/pre_tool_use.rs:416) has one production caller: `get_index_status` at crates/mcp-server/src/hook_handlers/pre_tool_use.rs:477, which loads index data and falls back to a not-indexed `IndexStatus` when none exists (crates/mcp-server/src/hook_handlers/pre_tool_use.rs:469). The other two call sites are tests, both passing the fixture path `[path]/projects/example-app/src`: crates/mcp-server/src/hook_handlers/pre_tool_use.rs:4119 and crates/mcp-server/src/hook_handlers/pre_tool_use.rs:4138.

Call sites

FileLineCode
crates/mcp-server/src/hook_handlers/pre_tool_use.rs477index_status_from_data(folder_path, data)
crates/mcp-server/src/hook_handlers/pre_tool_use.rs4119let status = index_status_from_data("[path]/projects/example-app/src", &data);
crates/mcp-server/src/hook_handlers/pre_tool_use.rs4138let status = index_status_from_data("[path]/projects/example-app/src", &data);

Read next

  • crates/mcp-server/src/hook_handlers/pre_tool_use.rs:472-592: Definition of index_status_from_data and its IndexStatus return type
Checks: 5 of 5 passed
  • index_read: 2 of 2 searches answered; 44 hits in 14 files.
  • citations: Every citation is to an item found in the index or the workspace.
  • answer_size: About 143 tokens; the limit is 600.
  • call_sites: 3 call site(s) of index_status_from_data, from every line that holds its name.
  • workspace_context: 0 context items from ContextStream for this job, about 245 tokens.
  • The paths under [path] are a test fixture's home directory in the repository's own tests.
Task
cf54234c-6b6c-4d63-8cc3-0d0daf973193

Live run on ContextStream’s task rail

Where does the server work out which files are exact-search only?

Agent
oracle@0.2.0
Outcome
succeeded
Ran
2026-10-27 17:18 UTC
Took
8.0 s
Cost
Free

ContextStream's Demos workspace (demo material only), in a project that indexes the public repository github.com/contextstream/mcp-server at v1.0.34.

Input
{
  "project_id": "6aad1065-8086-4114-931e-1b16c1d9b234",
  "question": "Where does the MCP server work out which files are exact-search only, and where does index_status show it?"
}

Result

Answer

The MCP server reads index_status in crates/mcp-server/src/hook_handlers/pre_tool_use.rs: `index_status_from_data` (crates/mcp-server/src/hook_handlers/pre_tool_use.rs:416) parses the status payload and `get_index_status` (crates/mcp-server/src/hook_handlers/pre_tool_use.rs:469) fetches it. The exact-search-only breakdown is rendered by `format_exact_search_only` in the mcp-tools project domain (context:251b0446-efae-45f8-860d-b31c61a3d34a), which reads the `exact_search_only` summary's `files` and `by_reason` counts; a test pins that shape (context:4e878000-feda-4f88-9b6e-4a8b4ad3d510). The hits do not show the code that classifies individual files as exact-search-only — that computation is absent from the candidates; crates/mcp-tools/src/domains/search.rs only covers exact-search budget/failure handling (crates/mcp-tools/src/domains/search.rs:2807).

Read next

  • crates/mcp-server/src/hook_handlers/pre_tool_use.rs:379-409: Parses index_status payload; get_index_status fetches it nearby
  • crates/mcp-tools/src/domains/search.rs:7307-7337: Exact-search failure/budget handling, not file classification
  • crates/mcp-server/src/hook_handlers/git_post_checkout.rs:1-26: Ranked 2 by the index search.
Checks: 4 of 4 passed
  • index_read: 7 of 7 searches answered; 265 hits in 91 files.
  • citations: Every citation is to an item found in the index or the workspace.
  • answer_size: About 216 tokens; the limit is 600.
  • workspace_context: 5 context items from ContextStream for this job, about 234 tokens.
  • A harder question: the answer says plainly which part it could not find (confidence 0.45).
Task
549b4aa8-274d-4b36-a388-318768d04e03

Example inputs

Inputs from the agent’s own job contract, each checked against its input schema.

The file and lines that define it, and what to read next.

{
  "question": "Where is the task route allow list defined?",
  "project_id": "00000000-0000-4000-8000-0000000000a1"
}
ss run oracle question='Where is the task route allow list defined?' project_id=00000000-0000-4000-8000-0000000000a1

Every call site found by exact search, with file and line.

{
  "question": "What calls `authorize_task_route`?",
  "project_id": "00000000-0000-4000-8000-0000000000a1"
}
ss run oracle question='What calls `authorize_task_route`?' project_id=00000000-0000-4000-8000-0000000000a1

A cited walk through the main steps, and the ranges to read.

{
  "question": "How does a model call flow from the worker to the proxy?",
  "project_id": "00000000-0000-4000-8000-0000000000a1"
}
ss run oracle question='How does a model call flow from the worker to the proxy?' project_id=00000000-0000-4000-8000-0000000000a1

The places to change, best first.

{
  "question": "Where do I add a new scope for task tokens?",
  "project_id": "00000000-0000-4000-8000-0000000000a1"
}
ss run oracle question='Where do I add a new scope for task tokens?' project_id=00000000-0000-4000-8000-0000000000a1

Declined: changing code is not Code Oracle's job.

{
  "question": "Fix the retry bug in the uploader",
  "project_id": "00000000-0000-4000-8000-0000000000a1"
}
ss run oracle question='Fix the retry bug in the uploader' project_id=00000000-0000-4000-8000-0000000000a1

Input

FieldTypeWhat it is
questionstring, requiredOne question about the code: where something is, what calls something, how something flows, or where to make a change.
project_idstring, requiredThe indexed project (repository) to answer from.
pathsstring[]Path prefixes to prefer, for example ["crates/api/src"].

Where it runs

Model agent. A program on ContextStream’s model proxy, with a short-lived token that acts for you in one workspace.

It may run for up to 1 min before it is stopped.

ContextStream scopeWhat it lets a run do
search.readSearch the workspace’s code, by keyword and by meaning.
memory.readSearch memory, and read decisions and lessons.
context.readRead the run’s context pack from the workspace.
model.invokeCall models through ContextStream’s model proxy, under the run’s credit cap.

Evals

202 of 212 offline eval cases pass for version 0.2.0 (baseline saved 2026-10-27). Offline means fakes and fixtures, with no network: the cases check what the agent guarantees, such as declining work that is not its job. They are not measurements of live runs; the live runs above are.