pjt222/agent-almanac/plan-release-cycle
- community
Checking that the SKILL.md file is still in the repository.
Plan software release cycle: milestones, feature freezes, RCs, go/no-go. Calendar-based + feature-based strategies. Use → start major/minor release planning, ad-hoc → structured cadence, multi-team release coord, regulated quality gates, first public v1.0.0.
Install
npx vskill@latest install pjt222/agent-almanac/plan-release-cycleSKILL.md
Plan Release Cycle
Plan structured release cycle: pick strategy (calendar or feature), set milestones, freeze criteria, RCs, go/no-go, rollback. Output: RELEASE-PLAN.md artifact guiding team dev → release.
Use When
- Major/minor release planning starts
- Ad-hoc → structured release cadence
- Multi-team or multi-component release coord
- Quality gates + criteria for regulated project
- First public release (v1.0.0)
In
- Required: Target version (e.g., v2.0.0)
- Required: Release date or window
- Required: Planned features/scope (backlog, roadmap, desc)
- Optional: Team size + availability
- Optional: Strategy pref (calendar or feature)
- Optional: Regulatory/compliance reqs
- Optional: Past release velocity or cycle data
Do
Step 1: Determine Strategy
Two primary strategies:
Calendar-based (time-boxed):
- Fixed schedule (e.g., 4 wks, quarterly)
- Not-ready features → next release
- Predictable for users + downstream
- Best for: libraries, frameworks, tools w/ external consumers
Feature-based (scope-driven):
- Release when defined scope done
- Date adjusts to scope
- Risk: scope creep, indefinite delays
- Best for: internal tools, first releases, major rewrites
Most projects → hybrid: target date + defined scope, 1-2 week buffer. Scope not met by buffer → defer remaining.
Doc strategy choice w/ rationale.
→ Strategy doc'd w/ rationale matching project context.
If err: team can't agree → default calendar-based w/ feature-priority list. Time-box forces prioritization.
Step 2: Define Milestones
Phases w/ target dates:
Typical durations:
- Development: 50-70% of cycle
- Stabilization: 15-25% of cycle
- RC testing: 10-20% of cycle
→ Milestone table: dates, durations, descriptions per phase.
If err: timeline too compressed (stabilization < 1 wk) → extend release date or reduce scope. Never skip stabilization.
Step 3: Set Feature Freeze Criteria
Define "feature freeze" for this release:
P0 → blocks release. P1 → in if ready. P2 → deferred w/o delay.
→ Freeze rules doc'd w/ exception process + prioritized list.
If err: P0 at risk of missing freeze → escalate immediately. Options: extend dev phase, split feature smaller, defer to point release (v2.0.1).
Step 4: Plan RC Process
How RCs are produced + tested:
→ RC process doc'd: tagging convention, distribution, testing checklist, escalation.
If err: RC skipped (release pressure) → doc the risk. Untested releases → higher rollback prob.
Step 5: Define Go/No-Go Checklist
Criteria before release approval:
→ Go/no-go checklist w/ clear pass/fail + decision rules.
If err: meeting → no-go → ID blockers, assign owners, new target (typically 1-2 wks later), update plan.
Step 6: Document Rollback Plan
Roll back if release causes critical prod issues:
Write complete plan → RELEASE-PLAN.md or RELEASE-PLAN-v2.0.0.md.
→ Rollback plan doc'd: triggers, procedure, comm template, point release policy. Complete RELEASE-PLAN.md written.
If err: rollback not feasible (DB migration applied) → doc forward-fix instead. Every release needs recovery path.
Check
- [ ] Strategy (calendar/feature/hybrid) doc'd w/ rationale
- [ ] Milestone table all phases: dev, freeze, stabilization, RC, release
- [ ] Freeze criteria w/ allowed/disallowed change types
- [ ] Feature priority list (P0 / P1 / P2)
- [ ] RC process doc'd: tagging, distribution, testing, escalation
- [ ] Go/no-go has "must pass" + "should pass" sections
- [ ] Rollback plan: triggers, procedure, comm template
- [ ] RELEASE-PLAN.md created + saved
- [ ] Timeline realistic (stabilization ≥ 15% of cycle)
Traps
- No stabilization phase: Direct dev → release. Even 3-day stabilization catches issues active dev masks.
- Scope creep after freeze: "Just one more feature" → resets testing, regression risk.
- Ignore P0 risks: Not escalating early. Earlier scope adjust = less timeline disruption.
- Skip RC for "small" releases: Even minor → benefit from 1 RC. Day of RC < post-release hotfix.
- No rollback plan: Assume success. Every plan answers "what if this goes wrong?" before publish.
- Calendar pressure > quality: Release on promised date despite failing go/no-go. Delayed release = minor inconvenience; broken release = trust violation.
→
apply-semantic-versioning— determine version for planned releasemanage-changelog— maintain changelog feeding release notesplan-sprint— sprint planning within dev phasedraft-project-charter— may define release roadmap + success criteriagenerate-status-report— track progress vs milestones
Security
- Security scanPassed, 95 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingi18n/caveman-ultra/skills/plan-release-cycle/SKILL.md
Popularity
- Repository stars
- 15
- Repository forks
- 1
- Repository npm downloads, 7 days
- 22
- 7-day trend score
- 25
Stars and forks are for the whole GitHub repository, not this skill alone. Downloads are for the repository's npm package. Refreshed 15 May 2026, 5 months ago.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/pjt222/agent-almanac/plan-release-cycle)