Thales Maciel ab02ae46c7 Add model-native workspace file operations

Remove shell-escaped file mutation from the stable workspace flow by adding explicit file and patch tools across the CLI, SDK, and MCP surfaces.

This adds workspace file list/read/write plus unified text patch application, backed by new guest and manager file primitives that stay scoped to started workspaces and /workspace only. Patch application is preflighted on the host, file writes stay text-only and bounded, and the existing diff/export/reset semantics remain intact.

The milestone also updates the 3.2.0 roadmap, public contract, docs, examples, and versioning, and includes focused coverage for the new helper module and dispatch paths.

Validation:
- uv lock
- UV_CACHE_DIR=.uv-cache make check
- UV_CACHE_DIR=.uv-cache make dist-check
- real guest-backed smoke for workspace file read, patch apply, exec, export, and delete

2026-03-12 22:03:25 -03:00

2.9 KiB

Raw Blame History

LLM Chat Ergonomics Roadmap

This roadmap picks up after the completed workspace GA plan and focuses on one goal:

make the core agent-workspace use cases feel trivial from a chat-driven LLM interface.

Current baseline is 3.2.0:

the stable workspace contract exists across CLI, SDK, and MCP
one-shot pyro run still exists as the narrow entrypoint
workspaces already support seeding, sync push, exec, export, diff, snapshots, reset, services, PTY shells, secrets, network policy, and published ports
stopped-workspace disk tools now exist, but remain explicitly secondary

What "Trivial In Chat" Means

The roadmap is done only when a chat-driven LLM can cover the main use cases without awkward shell choreography or hidden host-side glue:

cold-start repo validation
repro plus fix loops
parallel isolated workspaces for multiple issues or PRs
unsafe or untrusted code inspection
review and evaluation workflows

More concretely, the model should not need to:

patch files through shell-escaped printf or heredoc tricks
rely on opaque workspace IDs without a discovery surface
consume raw terminal control sequences as normal shell output
choose from an unnecessarily large tool surface when a smaller profile would work

Locked Decisions

keep the workspace product identity central; do not drift toward CI, queue, or runner abstractions
keep disk tools secondary and do not make them the main chat-facing surface
prefer narrow tool profiles and structured outputs over more raw shell calls
every milestone below must update CLI, SDK, and MCP together
every milestone below must also update docs, help text, runnable examples, and at least one real smoke scenario

Milestones

Completed so far:

3.2.0 added model-native workspace file * and workspace patch apply so chat-driven agents can inspect and edit /workspace without shell-escaped file mutation flows.

Expected Outcome

After this roadmap, the product should still look like an agent workspace, not like a CI runner with more isolation.

The intended model-facing shape is:

one-shot work starts with vm_run
persistent work moves to a small workspace-first contract
file edits are structured and model-native
workspace discovery is human and model-friendly
shells are readable in chat
the five core use cases are documented and smoke-tested end to end

2.9 KiB Raw Blame History