Roadmap

Direction In One Sentence

gig is moving from a capable local release helper into a remote-first, zero-config-first release audit CLI.

What Is Already Shipping

The current codebase already delivers:

  • ticket-aware commit discovery across repositories
  • release verification and risk hints
  • Markdown and JSON release packets
  • remote inspection for GitHub, GitLab, Bitbucket, Azure DevOps, and remote SVN
  • current-checkout remote inference from Git origin
  • high-confidence protected-branch inference with explicit fallback when topology is ambiguous
  • local Git and SVN fallback flows
  • reusable projects, with workarea kept as a compatibility alias
  • a guided gig front door
  • optional DeerFlow-backed ticket, release, and conflict briefings with saved follow-up sessions and richer release evidence
  • structured diagnostics for auth and topology support traces

What Comes Next

1. Sharper First-Run Experience

Priority:

  • reduce friction from install to first useful audit
  • let users run gig ABC-123, gig verify ABC-123, and gig packet ABC-123 from inside the repo without typing --repo
  • improve repository discovery and project reuse

2. Stronger Remote Audit Depth

Priority:

  • improve cross-branch and cross-repo ticket evidence
  • strengthen follow-up fix detection
  • make safe, warning, and blocked verdicts easier to trust at a glance
  • raise more providers from basic release evidence to deep release evidence

3. Better Project And Console UX

Priority:

  • cleaner multi-project switching
  • stronger summary-first terminal output
  • faster keyboard-driven navigation for repeated use

4. Better Release Evidence

Priority:

  • richer release packets and JSON contracts
  • stronger audit bundles for QA, release, and client-facing stakeholders
  • keep deep release evidence strong across GitHub, GitLab, and Azure DevOps while closing the remaining Bitbucket parity gap
  • optional AI explanations that stay grounded in deterministic evidence

5. Controlled Expansion

Priority:

  • fill the remaining provider and enterprise gaps without making the product config-heavy again
  • add guarded write actions only after the read-only audit path is strong enough

What Should Not Lead

These are useful, but they should not become the front door:

  • config-first onboarding
  • local-workspace-first storytelling
  • exposing engine internals before user intent is clear
  • adding more required branch flags where source control can answer directly

Product Rule

Safe release decisions matter more than clever automation.