THE FIELD GUIDE / 06Coding & research

Follow the question all the way into the work.

Explore a codebase, investigate a failure, research sources and create useful artifacts.

5 capabilities3 practical walkthroughs
An ivory coding workspace with a deep indigo screen
ONE LITTLE REQUESTRead this project, reproduce the failing test, and fix the smallest relevant part.
JARVIS / IN EVERYDAY LIFEIllustrated possibilities. Your tools do the work.
01 / THE POSSIBILITIES

A little more
within reach.

Some requests need a chain of work: read the project, understand the failure, inspect a source, make a careful edit and check the result. The high-level agent brings file, terminal, web and delegation tools to those longer tasks.

See it in your day
01

Repository exploration

Read files, search contents and trace related code before proposing an implementation or explanation.

02

Terminal & code execution

Run permitted commands and inspect output. Code mode offers a persistent Python execution workspace for multi-step tasks.

03

Diagnostics & debugging

Use connected VS Code diagnostics or the Python diagnostic tools backed by installed ty, pyright or mypy.

04

Specialist delegation

Delegate exploration, planning or general work to configured subagents and use their findings within the larger task.

05

Web research & writing

Search sources, extract page text, compare evidence and write a summary or local artifact through the configured tools.

02 / PUT IT INTO YOUR DAY

Ordinary requests.
Useful little transformations.

Follow the thought from the first sentence to the result. Make the names, files, and timing your own.

THE EVERYDAY / WORKFLOW 01

Move from an error to a checked change

The error is visible but its cause crosses several files.

An illustrated example of move from an error to a checked change
AN ILLUSTRATED WORKFLOW
START WITH THIS
Read this project, reproduce the failing test, and fix the smallest relevant part.
  1. 01

    Start with the context

    Jarvis resolves the project and reads the relevant implementation, configuration and failure context.

  2. 02

    Let the work unfold

    The task agent runs the permitted reproduction, makes a targeted edit and can use installed diagnostics to refine the fix.

  3. 03

    Check the result

    It runs the relevant check again and reports the actual result. Review the diff and any remaining failure before merging or releasing.

THE RESULT TO LOOK FOR

A local change with specific validation evidence if the environment supports it. A plausible explanation alone is not a passing test.

THE EVERYDAY / WORKFLOW 02

Turn several sources into a useful brief

You need a decision you can revisit, rather than another collection of open tabs.

An illustrated example of turn several sources into a useful brief
AN ILLUSTRATED WORKFLOW
START WITH THIS
Compare these three documentation pages and save a short implementation brief with source links.
  1. 01

    Start with the context

    The agent retrieves the specified sources through available web or browser tools.

  2. 02

    Let the work unfold

    It compares the relevant claims, notes missing access or conflicting information, and organizes the findings around your question.

  3. 03

    Check the result

    It writes the requested brief to a file. Open it to verify the links, conclusions and any assumptions before using the recommendation.

THE RESULT TO LOOK FOR

A saved research artifact grounded in retrieved sources. Access restrictions and unavailable pages can limit the comparison.

THE EVERYDAY / WORKFLOW 03

Give a difficult bug a better starting point

An error involves the desktop, the active project and editor diagnostics; retelling it would lose context.

An illustrated example of give a difficult bug a better starting point
AN ILLUSTRATED WORKFLOW
START WITH THIS
Capture this error so we can investigate it.
  1. 01

    Start with the context

    The desktop workflow collects the available current context and, when enabled, a screenshot.

  2. 02

    Let the work unfold

    It adds connected editor diagnostics and saves the captured materials in a local problem bundle.

  3. 03

    Check the result

    The result names missing components. Ask the task agent to inspect the bundle and investigate the code using the evidence actually captured.

THE RESULT TO LOOK FOR

A local evidence bundle to support debugging. A capture is the beginning of an investigation, not a diagnosis by itself.

03 / MAKE THE CONNECTIONS

A little preparation.
A lot of possibility.

An illustrated setup sheet: download, configure, say helloOpen the installation guide
01

Configure the task model

Choose a supported provider and its credentials. A local model server must be installed and running separately.

uv run jarvis model
02

Prepare the project runtime

Install the dependencies, test runner and command-line tools needed by the project being worked on. Jarvis uses the environment you supply.

03

Connect diagnostics

The VS Code adapter exposes supported context and diagnostics. Python language checks require an installed ty, pyright or mypy backend.

uv run jarvis desktop setup-editor --install
04 / A CLEAR PICTURE

Good to know
before you begin.

Access follows configuration

Tool permissions, agent profiles and the available environment determine what can run. A skill or requested outcome does not grant credentials or bypass access rules.

Separate stages of delivery

Editing, testing, committing, publishing and deploying are separate actions. A successful local check does not establish that a remote deployment or account operation worked.

05 / A LITTLE MORE CONTEXT

The natural next questions.

Workflow images are generated illustrations, not screenshots of a connected assistant. Availability depends on your configured tools and providers.

FOLLOW YOUR CURIOSITY

A few more possibilities.

Back to all features