Skip to main content

Roadmap

This roadmap is directional. It explains product sequencing, not date commitments.

Available now: local task flow, repo context, and agent adapters​

The local CLI provides:

  • ds task creates bounded task workspaces with packed context;
  • ds apply emits the unambiguous next one-slice prompt without launching an agent;
  • ds task "<goal>" --quick creates a compact one-off task workspace;
  • ds task slice add --after <slice> --reason <gate> creates A01-1 style follow-up slices after improve or rework;
  • ds map, ds recent, and ds find recover local repo context;
  • ds map indexes the repository before identifying system boundaries;
  • ds recent merges overlapping recent work and demotes generic maintenance noise;
  • cold indexing reports progress while preserving quiet and JSON result safety;
  • ds init can write small repo-local files for Codex, Cursor, Claude, or Windsurf that route agents back through ds task and ds apply.

The current generated entry points are $ds-task / $ds-apply for generic agent skill files and /ds-task / /ds-apply for Cursor, Claude, and Windsurf-style files.

Shipped Experimental: workspace coordination​

Many teams work across more than one repo. Experimental workspace commands are available under ds workspace ... and the ds ws shortcut:

  • ds workspace init creates an umbrella workspace manifest;
  • ds workspace change create records workspace-level change intent;
  • ds workspace slice create creates linked repo-local task slices;
  • ds workspace trace follows known workspace changes and repo task IDs.

Workspace coordination is still experimental. Prefer normal repo-local ds task, ds apply, and ds task checkpoint for single-repo work.

Short-Term: durable wrappers for existing intent​

Repos already have useful PRDs, RFCs, ADRs, plans, specs, and agent notes. DevSpecs can already index and retrieve those artifacts in place through ds scan, ds map, ds recent, ds find, and ds context.

ds adopt is planned for the next step: optional wrapper artifacts that make older intent files explicitly addressable, metadata-rich, and easy to version without asking teams to rewrite them.

Expected shape:

  • detect candidate intent artifacts during map/find/scan flows;
  • show a dry run before writing anything;
  • create thin wrapper artifacts for addressability and metadata;
  • leave original files untouched by default;
  • make wrappers easy to version when they should become durable team context.

Short-Term: richer agent wrappers​

Initial adapter files are shipped through ds init. The next step is polish:

  • clearer tool-specific recipes for task creation and apply prompts;
  • better generated instructions for one-target work;
  • safer defaults for agents that restart shells or run inside sandboxes;
  • future local MCP adapter over bounded commands such as ds apply, ds task checkpoint, ds task status, ds map, ds recent, and ds find.

The CLI remains the source of truth. Integrations should wrap it, not create a parallel workflow.

Mid-Term: workspace hardening​

Workspace support should become less experimental only after the CLI proves the storage and handoff model across real umbrella repos.

Likely areas:

  • clearer cross-repo task and intent views;
  • stronger workspace-level config;
  • clearer storage choices for personal versus team artifacts;
  • safer series naming when multiple people create task tracks.

Long-Term: research-backed context quality​

We are actively researching better ways to identify the right source, test, and intent context from local evidence. The bar for shipping this into DevSpecs is high: improvements should be deterministic, explainable, and useful without requiring cloud sync.

Expected shape:

  • better context packing and discovery from local repository evidence;
  • clearer positive and negative evidence for why files were included or excluded;
  • public writeups when the results are reproducible and useful.