gmgauthier/grokkit
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases 10 Wiki Activity
4f40c7e8d7
grokkit/todo/completed/3-new-feature-suggestions.md
Greg Gauthier 6e3a52728e
All checks were successful
CI / Test (push) Successful in 30s
Details
CI / Lint (push) Successful in 26s
Details
CI / Build (push) Successful in 20s
Details
chore(todo): complete feature suggestions TODO and queue individual tasks 
- Update CI workflow runner from ubuntu-latest to ubuntu-gitea for auto-complete-todo.
- Move 3-new-feature-suggestions.md to completed/.
- Create queued TODOs for testgen, changelog, and profile features.
2026-03-02 19:33:34 +00:00

2.0 KiB
Raw Blame History

3 New AI-Enhanced Feature Suggestions for Grokkit

These suggestions build on existing CLI strengths (chat/edit/lint/agent) to further streamline Go development workflows.

1. grokkit testgen [path...]

Description: Generate comprehensive unit tests for Go files/packages using AI analysis of source code.

Benefits:

  • Boost test coverage from current ~72% baseline automatically.
  • Enforces modern table-driven tests with t.Parallel() matching codebase style (e.g., internal/version/version_test.go).

High-level implementation:

  • Extract funcs/structs via go/ast or prompt with code snippets.
  • Prompt Grok: "Generate table-driven tests for this Go code."
  • Output *_test.go, create .bak, preview/apply like edit.
  • Verify: go test -v ./... post-generation.

CLI example:

grokkit testgen internal/grok/client.go

2. grokkit changelog

Description: AI-generated CHANGELOG.md updates from git history (commits/tags).

Benefits:

  • Automates semantic release notes (feat/fix/docs/etc.).
  • Integrates with TODO workflow for release PRs.

High-level implementation:

  • Fetch git log --pretty=format:%s%n%b --since=<last-tag>.
  • Prompt Grok: "Categorize into CHANGELOG sections."
  • Append/update CHANGELOG.md, preview/commit.

CLI example:

grokkit changelog --since=v0.1.3
grokkit changelog --commit  # Stage + grokkit commit

3. grokkit profile

Description: Run Go benchmarks/pprof, get AI-suggested performance optimizations.

Benefits:

  • Tune CLI hotspots (e.g., API streaming, git subprocesses).
  • Proactive perf improvements without manual profiling.

High-level implementation:

  • go test -bench=. -cpuprofile=prof.out ./cmd.
  • Parse pprof data (stdlib pprof or text), prompt Grok: "Optimize these hotspots."
  • Suggest edits via agent or edit --preview.

CLI example:

grokkit profile ./cmd/agent.go
# Outputs: hotspots, suggestions, optional auto-fixes

These features align with minimal deps, modern Go, and existing patterns (Cobra cmds, grok client, git integration).