ContribAI Sprint & Release Plan

Current Version: v6.0.0 | Date: 2026-04-11 | Status: Active Development

Executive Summary

ContribAI is at v6.0.0 with a mature Rust codebase (587+ tests, 67+ .rs files, ~30k LOC). The project has completed 16 sprints covering core pipeline, CLI/TUI, AST analysis, memory systems, security, plugins, and i18n.

This plan outlines 24 sprints organized into 8 releases (v6.1.0 → v7.0.0), targeting Q2-Q4 2026. Each release is triggered via gh tag push to automate GitHub Actions CD.

Release Strategy

Tagging & CD Workflow

# Each release triggered by:
git tag -a v6.1.0 -m "Release v6.1.0 — <feature summary>"
git push origin v6.1.0

# GitHub Actions (release.yml) auto-builds:
# - Linux x86_64
# - Windows x86_64  
# - macOS x86_64 (Intel)
# - macOS aarch64 (Apple Silicon)

Release Cadence

Minor releases (v6.x.0): Every 2-3 sprints (~2-3 weeks)
Patch releases (v6.x.1): Bug fixes only (as needed)
Major releases (v7.0.0): Breaking changes or milestone features

Sprint Details

Release v6.1.0 — Quality & Performance Polish

Target: Sprint 17-18 | Timeline: 2 weeks | Tag: v6.1.0

Sprint 17: Code Quality & Dead Code Removal

Priority: High | Estimate: 3-4 days

Task	Files	Details
Remove Session dead code	`cli/commands/session.rs`, `cli/mod.rs`	Wire Session command or remove module
Wire Copilot provider	`llm/provider.rs`, `llm/copilot.rs`	Add Copilot to `create_llm_provider()` factory
Fix clippy warnings (6)	Various	Remove unused imports, derive Default, fix unwrap pattern
Clean test helper warnings	`tests/common/mock_*.rs`	Remove 14 unused helper functions or mark `#[allow(dead_code)]`
Framework detection stub	`analysis/analyzer.rs:136`	Implement basic framework detection from imports

Deliverables:

0 clippy warnings (strict lint enforcement)
Session command either functional or removed
Copilot provider usable via config
Framework detection returns non-empty HashSet for Django/Flask/React projects

Acceptance Criteria:

cargo clippy -- -D warnings  # Zero warnings
cargo test                    # All 587+ tests pass

Sprint 18: Performance & Dependency Updates

Priority: High | Estimate: 3-4 days

Task	Files	Details
Update tree-sitter version mismatch	`Cargo.toml`	Align core `tree-sitter 0.24` with grammars (currently `0.23`)
Update tower 0.4 → 0.5	`Cargo.toml`, `web/mod.rs`	Compatible with axum 0.7
Add Rust dependabot	`.github/dependabot.yml`	Automated Rust dependency updates
Dependency audit	`Cargo.toml`	Review all 60+ deps, update outdated, remove unused
Binary size optimization	`Cargo.toml` (profile.release)	Benchmark current size, apply further optimizations
Benchmark suite	`benches/pipeline.rs`	Add criterion benchmarks for hot paths

Deliverables:

All tree-sitter grammars at 0.24
dependabot PRs for Rust deps
Binary size < 4.5MB stripped
Baseline performance benchmarks

Acceptance Criteria:

cargo bench                    # Benchmarks run successfully
gh dependabot list             # Rust repo covered

Release v6.2.0 — Documentation & Developer Experience

Target: Sprint 19-20 | Timeline: 2 weeks | Tag: v6.2.0

Sprint 19: Documentation Overhaul

Priority: High | Estimate: 3 days

Task	Files	Details
Update CHANGELOG.md	`CHANGELOG.md`	Add v5.18.0-v6.0.0 entries (currently stops at v5.17.0)
Update README.md badges	`README.md`	Fix test count: 418 → 587+
Sync AGENTS.md version refs	`AGENTS.md`	Update v5.8.1 → v6.0.0 throughout
Update roadmap docs	`docs/project-roadmap.md`	Mark v5.18.0-v6.0.0 complete, add 2026 Q2-Q4 plans
Add RUNBOOK.md emergency procedures	`RUNBOOK.md`	Expand with common failure modes + recovery
Add CONTRIBUTING.md dev guide	`CONTRIBUTING.md`	How to add commands, tests, run CI locally

Deliverables:

All docs reference v6.0.0+
Accurate README badges
Complete changelog through v6.0.0
Developer onboarding guide

Sprint 20: Developer Experience Improvements

Priority: Medium | Estimate: 4 days

Task	Files	Details
Add `make dev` target	`Makefile`	One-command setup: build + test + lint
Add pre-commit hooks	`.pre-commit-config.yaml`	Auto-run cargo fmt, clippy, test on commit
Add justfile alternative	`justfile`	Modern command runner (optional, alongside Makefile)
Improve error messages	`core/error.rs`	User-friendly errors with suggestions (not just "failed")
Add `--verbose` flag globally	`cli/mod.rs`	Enable tracing output for debugging
Add shell completions	`cli/mod.rs`	Generate bash/zsh/fish/pwsh completions via clap_complete

Deliverables:

make dev runs successfully from fresh clone
Pre-commit hooks enforce quality gates
Shell completions available
Error messages actionable (e.g., "Token invalid. Run contribai login to re-authenticate")

Release v6.3.0 — Advanced Analysis & Generation

Target: Sprint 21-22 | Timeline: 2-3 weeks | Tag: v6.3.0

Sprint 21: Multi-Hop Import Resolution

Priority: High | Estimate: 4-5 days

Task	Files	Details
Extend import resolution depth	`analysis/ast_intel.rs`	1-hop → 2-hop (cap at 50 symbols)
Add cross-language imports	`analysis/ast_intel.rs`	Detect FFI bindings (Python→Rust, JS→WASM)
Import resolution cache	`analysis/ast_intel.rs`	Cache resolved imports in SQLite (TTL: 24h)
Symbol usage frequency	`analysis/ast_intel.rs`	Track which symbols are most referenced
Pipeline integration	`orchestrator/pipeline.rs`	Use import graph for smarter file selection

Deliverables:

2-hop import resolution (50-symbol cap)
Cross-language FFI detection
Import cache reduces analysis time by 30%+
Pipeline prioritizes files with high import centrality

Sprint 22: LLM Generation Improvements

Priority: High | Estimate: 4 days

Task	Files	Details
Structured output parsing	`generator/engine.rs`	Enforce JSON schema on LLM responses
Self-correction loop	`generator/self_review.rs`	Re-generate if validation fails (max 2 retries)
Multi-file dependency ordering	`generator/engine.rs`	Ensure edits respect file dependencies (order matters)
Diff quality scoring	`generator/validation.rs`	Score generated diffs before submission
Reduce false positives	`analysis/triage.rs`	Tune 12-signal scoring to reduce noise

Deliverables:

90%+ LLM responses parse correctly (up from ~75%)
Self-correction reduces invalid PRs by 40%
Multi-file PRs respect dependency order
Diff quality gate blocks low-quality submissions

Release v6.4.0 — Enterprise Features

Target: Sprint 23-24 | Timeline: 3 weeks | Tag: v6.4.0

Sprint 23: PostgreSQL & Redis Support

Priority: High | Estimate: 5 days

Task	Files	Details
Database abstraction trait	`orchestrator/db.rs`	Abstract SQLite/PostgreSQL behind common interface
PostgreSQL implementation	`orchestrator/db_postgres.rs`	sqlx + tokio-postgres async driver
Feature flag for DB backend	`Cargo.toml`	`features = ["sqlite", "postgres"]`
Redis cache layer	`llm/cache.rs`, `orchestrator/memory.rs`	Optional Redis for distributed caching
Migration system	`orchestrator/migrations.rs`	Schema versioning + auto-migrate on startup

Deliverables:

Config-selectable SQLite/PostgreSQL backend
Redis cache for LLM responses (shared across instances)
Automatic schema migrations
Feature-gated dependencies (no bloat for SQLite-only users)

Config Example:

database:
  backend: postgresql  # or sqlite
  url: "postgresql://user:pass@localhost/contribai"
  
cache:
  backend: redis  # or sqlite
  url: "redis://localhost:6379"

Sprint 24: Observability & Monitoring

Priority: Medium | Estimate: 3 days

Task	Files	Details
OpenTelemetry tracing	`core/tracing.rs`	Distributed tracing spans for pipeline, LLM, GitHub API
Prometheus metrics expansion	`web/mod.rs`	Add 20+ new metrics (latency, error rates, cache hit rate)
Grafana dashboard	`docs/grafana-dashboard.json`	Pre-built dashboard JSON for import
Health check endpoint	`web/mod.rs`	`/health` endpoint for Kubernetes liveness probes
Audit log export	`core/events.rs`	Stream JSONL events to external systems (S3, ELK)

Deliverables:

Full OpenTelemetry integration
30+ Prometheus metrics
Grafana dashboard template
Kubernetes-ready health checks

Release v6.5.0 — Plugin Ecosystem

Target: Sprint 25-26 | Timeline: 3 weeks | Tag: v6.5.0

Sprint 25: Plugin Framework v2

Priority: High | Estimate: 5 days

Task	Files	Details
WASM plugin runtime	`plugins/wasm.rs`	wasmtime for sandboxed plugin execution
Plugin manifest format	`plugins/manifest.rs`	YAML manifest with metadata, permissions, hooks
Plugin registry UI	`cli/tui.rs`	Tab for installed plugins, enable/disable, logs
Plugin marketplace schema	`docs/plugin-schema.json`	GitHub-based plugin registry format
Security sandboxing	`plugins/sandbox.rs`	Capability-based permissions (read/write/network)

Deliverables:

WASM-based plugin system (safe, sandboxed)
Plugin manifest with versioning
3 example plugins (Django analyzer, React linter, Go formatter)
Plugin CLI commands: plugin install/list/remove/info

Sprint 26: Pre-built Plugin Suite

Priority: Medium | Estimate: 4 days

Task	Details
Django plugin	Detect models, views, templates; generate migrations
React plugin	Component analysis, prop-type validation, hook optimization
Go plugin	Interface compliance, error handling patterns, benchmarks
Python async plugin	Detect blocking calls, recommend async/await
Security scanner plugin	CWE/SAST rules beyond built-in analyzer

Deliverables:

5 production-ready plugins
Plugin documentation + tutorial
Plugin security guidelines

Release v6.6.0 — Advanced PR Strategies

Target: Sprint 27-28 | Timeline: 2-3 weeks | Tag: v6.6.0

Sprint 27: PR A/B Testing & Learning

Priority: High | Estimate: 4 days

Task	Files	Details
PR title/body variants	`pr/manager.rs`	Test different PR formats (emoji vs no emoji, short vs detailed)
Merge rate prediction	`orchestrator/memory.rs`	ML-like scoring based on historical outcomes
Contribution type optimization	`orchestrator/memory.rs`	Auto-adjust strategy per repo (some prefer docs, some code)
Timing optimization	`scheduler/mod.rs`	Learn best times to submit PRs for visibility
Review response analysis	`pr/patrol.rs`	Parse reviewer tone/sentiment to improve future PRs

Deliverables:

A/B test PR formats
Predictive merge scoring (confidence %)
Repo-specific strategy adaptation
Optimal PR timing recommendations

Sprint 28: Multi-Repo Coordination

Priority: Medium | Estimate: 3 days

Task	Files	Details
Workspace detection	`github/discovery.rs`	Detect monorepos/workspaces (Nx, Cargo, Lerna)
Cross-repo PRs	`pr/manager.rs`	Coordinate changes across multiple repos
Shared dependency updates	`orchestrator/pipeline.rs`	Update library + all dependent projects
Batch PR grouping	`pr/manager.rs`	Group related changes into single PR

Deliverables:

Monorepo-aware contribution strategy
Cross-repo dependency updates
Intelligent PR batching

Release v6.7.0 — Security & Compliance

Target: Sprint 29-30 | Timeline: 2 weeks | Tag: v6.7.0

Sprint 29: Enterprise Security

Priority: High | Estimate: 4 days

Task	Files	Details
Real token encryption	`core/crypto.rs`	Replace XOR with `aes-gcm` crate (AES-256-GCM)
SSO/OAuth support	`github/client.rs`	GitHub App authentication flow
Audit mode	`cli/mod.rs`	Read-only analysis reporting (no PRs)
Compliance policies	`core/compliance.rs`	Enforce org-specific contribution policies
Secret scanning	`generator/validation.rs`	Detect leaked secrets in generated code

Deliverables:

Production-grade AES-256-GCM encryption
GitHub App OAuth flow (no PAT required)
Audit-only mode for compliance reviews
Built-in secret detection

Sprint 30: Governance & Reporting

Priority: Medium | Estimate: 3 days

Task	Files	Details
Activity reports	`cli/mod.rs`	Weekly/monthly contribution summaries
ROI metrics	`orchestrator/memory.rs`	Track value delivered (issues closed, bugs fixed)
Team dashboards	`web/mod.rs`	Multi-user activity tracking
Export to CSV/PDF	`web/mod.rs`	Downloadable reports for management

Deliverables:

Automated activity reporting
ROI tracking metrics
Team collaboration features

Release v7.0.0 — Full Agent Autonomy

Target: Sprint 31-40 | Timeline: 8-10 weeks | Tag: v7.0.0

Sprint 31-32: Agent Communication Protocol

Priority: High | Estimate: 8 days

Task	Files	Details
Agent message bus	`agents/bus.rs`	Pub/sub for inter-agent communication
Task delegation	`agents/registry.rs`	Agents can delegate subtasks to specialized agents
Shared context	`agents/context.rs`	Agents share analysis results, avoid duplicate work
Conflict resolution	`orchestrator/pipeline.rs`	Resolve conflicting agent recommendations
Agent orchestration UI	`cli/tui.rs`	Real-time agent activity monitoring

Deliverables:

5 agents communicate autonomously
Task delegation reduces completion time by 40%
Conflict resolution prevents contradictory PRs

Sprint 33-34: Self-Improvement System

Priority: High | Estimate: 8 days

Task	Files	Details
Outcome analysis	`orchestrator/memory.rs`	Analyze why PRs were merged/rejected
Strategy evolution	`orchestrator/pipeline.rs`	Auto-adjust parameters based on outcomes
Prompt optimization	`llm/formatter.rs`	A/B test prompts, track success rates
Model selection learning	`llm/router.rs`	Auto-choose best model per task type
Failure pattern database	`orchestrator/memory.rs`	Store failure modes for prevention

Deliverables:

Autonomous strategy tuning
Prompt performance tracking
Failure pattern recognition
Self-tuning model selection

Sprint 35-36: Multi-VCS Support

Priority: Medium | Estimate: 8 days

Task	Files	Details
VCS abstraction layer	`github/mod.rs` → `vcs/mod.rs`	Abstract Git/GitLab/Gitea operations
GitLab integration	`vcs/gitlab.rs`	Full GitLab API support
Gitea integration	`vcs/gitea.rs`	Self-hosted Gitea support
Gitee integration	`vcs/gitee.rs`	Chinese platform support
Cross-platform discovery	`github/discovery.rs`	Search across multiple VCS platforms

Deliverables:

Platform-agnostic core (works with GitLab, Gitea, Gitee)
Unified discovery across platforms
Platform-specific PR formatting

Sprint 37-38: Advanced Code Understanding

Priority: High | Estimate: 8 days

Task	Files	Details
Semantic code search	`analysis/search.rs`	Embedding-based code retrieval
Type inference	`analysis/ast_intel.rs`	Infer types without running code
Control flow analysis	`analysis/cfg.rs`	Build control flow graphs
Data flow analysis	`analysis/dfa.rs`	Track data dependencies
Impact prediction	`analysis/impact.rs`	Predict blast radius of changes

Deliverables:

Semantic code understanding
Type-aware generation
Impact analysis before PR submission

Sprint 39-40: Autonomous Operation Modes

Priority: High | Estimate: 8 days

Task	Files	Details
Goal-directed mode	`orchestrator/goal.rs`	"Improve test coverage to 80%" → autonomous execution
Continuous improvement	`orchestrator/pipeline.rs`	Continuously monitor and improve repos
Collaboration mode	`pr/manager.rs`	Work alongside human developers
Emergency response	`issues/solver.rs`	Rapid response to critical issues
Self-evaluation	`orchestrator/review.rs`	Agent reviews its own work before submission

Deliverables:

Goal-driven autonomous operation
Real-time collaboration features
Emergency issue triage + fixing
Self-evaluation quality gate

Sprint Execution Template

Pre-Sprint Checklist

Review previous sprint deliverables
Create sprint branch: git checkout -b sprint-XX
Update Cargo.toml version if needed
Run baseline tests: cargo test
Run clippy: cargo clippy -- -D warnings

During Sprint

Commit frequently with descriptive messages
Add tests alongside features
Update docs as you go
Run cargo fmt before commits

Post-Sprint Checklist

All tests pass: cargo test (expect 600+)
Zero clippy warnings: cargo clippy -- -D warnings
Update CHANGELOG.md with new entry
Update version in AGENTS.md, README.md if needed
Run integration tests if applicable
Create PR to main (if using sprint branches)

Release Checklist

Dependency Management

Critical Dependencies to Monitor

Dependency	Current	Target	Notes
tree-sitter	0.24 core, 0.23 grammars	0.24 all	Mismatch causes warnings
tower	0.4	0.5	Compatible with axum 0.7
axum	0.7	0.8 (when available)	Monitor for updates
rusqlite	0.31	Latest	Stable, low churn
tokio	1	Latest	Always stay current
tree-sitter-*	0.23	0.24	Align with core

Dependabot Schedule

# Add to .github/dependabot.yml
- package-ecosystem: "cargo"
  directory: "/crates/contribai-rs"
  schedule:
    interval: "weekly"
    day: "monday"
  open-pull-requests-limit: 10
  reviewers:
    - "tang-vu"

Risk Management

Technical Risks

Risk	Impact	Mitigation
LLM API cost increases	High	Improve cache, add budget controls
GitHub API rate limits	Medium	GraphQL batching, token rotation
Dependency breakage	Low	Strict version pinning, CI catches issues
Binary size growth	Medium	Regular profiling, tree-shaking audits
Test suite slowdown	Medium	Parallel test execution, flaky test monitoring

Schedule Risks

Risk	Impact	Mitigation
Sprint scope creep	High	Strict acceptance criteria, defer to next sprint
Complex features take longer	Medium	Break into smaller deliverables, use feature flags
Contributor availability	Low	Document thoroughly, async-friendly workflow

Success Metrics

Per-Release Targets

Metric	v6.1.0	v6.2.0	v6.3.0	v6.4.0	v6.5.0	v6.6.0	v6.7.0	v7.0.0
Test count	600+	620+	650+	680+	700+	720+	750+	800+
Clippy warnings	0	0	0	0	0	0	0	0
Binary size	<4.5MB	<4.5MB	<4.5MB	<5.0MB	<5.5MB	<5.5MB	<5.5MB	<6.0MB
CLI commands	42+	44+	45+	48+	50+	52+	54+	60+
LLM providers	7	7	7	7	7	7	8	10+
Languages	13	13	13	13	13	13	13	15+
Plugins	0	0	0	0	5+	8+	10+	15+
VCS platforms	1	1	1	1	1	1	4	4+

Quality Gates (Every Release)

✅ All tests pass
✅ Zero clippy warnings
✅ Zero cargo audit vulnerabilities
✅ Documentation updated
✅ CHANGELOG.md current
✅ Release binary tested on all 4 platforms

Quick Reference Commands

Daily Development

# Run tests
cargo test

# Lint
cargo clippy -- -D warnings

# Format
cargo fmt --all

# Build
cargo build --release

# Run specific test
cargo test hunt_integration

# Check for vulnerabilities
cargo audit

Release Process

# 1. Ensure clean state
git status
git diff HEAD

# 2. Run quality checks
cargo test && cargo clippy -- -D warnings && cargo audit

# 3. Update version
# Edit crates/contribai-rs/Cargo.toml
# Edit CHANGELOG.md

# 4. Commit
git commit -am "release: vX.Y.Z — <summary>"

# 5. Tag
git tag -a vX.Y.Z -m "Release vX.Y.Z — <summary>"

# 6. Push (triggers CD)
git push && git push origin vX.Y.Z

# 7. Monitor
gh run list --limit 5
gh release view vX.Y.Z

Emergency Rollback

# Delete tag
git tag -d vX.Y.Z
git push --delete origin vX.Y.Z

# Or revert commit
git revert HEAD

Appendix: Feature Request Backlog

Future Releases (Post-v7.0.0)

v7.1.0: Visual Studio Code extension
v7.2.0: Team collaboration features
v7.3.0: Advanced analytics dashboard
v8.0.0: Multi-agent swarms (10+ cooperating agents)

Nice-to-Have (Low Priority)

macOS Homebrew formula
Windows winget package
Docker image on GHCR
Nix flake support
contribai cloud service

Document Metadata

Created: 2026-04-11
Last Updated: 2026-04-11
Version: Planning for v6.1.0 → v7.0.0
Next Review: End of Sprint 17 (v6.1.0 release)
Owner: tang-vu@users.noreply.github.com

FilesExpand file tree

SPRINT_PLAN.md

Latest commit

History

SPRINT_PLAN.md

File metadata and controls

ContribAI Sprint & Release Plan

Executive Summary

Release Strategy

Tagging & CD Workflow

Release Cadence

Sprint Details

Release v6.1.0 — Quality & Performance Polish

Sprint 17: Code Quality & Dead Code Removal

Sprint 18: Performance & Dependency Updates

Release v6.2.0 — Documentation & Developer Experience

Sprint 19: Documentation Overhaul

Sprint 20: Developer Experience Improvements

Release v6.3.0 — Advanced Analysis & Generation

Sprint 21: Multi-Hop Import Resolution

Sprint 22: LLM Generation Improvements

Release v6.4.0 — Enterprise Features

Sprint 23: PostgreSQL & Redis Support

Sprint 24: Observability & Monitoring

Release v6.5.0 — Plugin Ecosystem

Sprint 25: Plugin Framework v2

Sprint 26: Pre-built Plugin Suite

Release v6.6.0 — Advanced PR Strategies

Sprint 27: PR A/B Testing & Learning

Sprint 28: Multi-Repo Coordination

Release v6.7.0 — Security & Compliance

Sprint 29: Enterprise Security

Sprint 30: Governance & Reporting

Release v7.0.0 — Full Agent Autonomy

Sprint 31-32: Agent Communication Protocol

Sprint 33-34: Self-Improvement System

Sprint 35-36: Multi-VCS Support

Sprint 37-38: Advanced Code Understanding

Sprint 39-40: Autonomous Operation Modes

Sprint Execution Template

Pre-Sprint Checklist

During Sprint

Post-Sprint Checklist

Release Checklist

Dependency Management

Critical Dependencies to Monitor

Dependabot Schedule

Risk Management

Technical Risks

Schedule Risks

Success Metrics

Per-Release Targets

Quality Gates (Every Release)

Quick Reference Commands

Daily Development

Release Process

Emergency Rollback

Appendix: Feature Request Backlog

Future Releases (Post-v7.0.0)

Nice-to-Have (Low Priority)

Document Metadata