retrospectives
Organizes developer feedback and reflections into a hierarchical system to drive project improvements.
Install
mkdir -p .claude/skills/retrospectives && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/11741" && unzip -o skill.zip -d .claude/skills/retrospectives && rm skill.zipInstalls to .claude/skills/retrospectives
Activation
This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.
The reflection and iterative improvement layer, managing project and session retrospectives.Key capabilities
- →Capture task-based retrospectives
- →Capture daily session-end reflections
- →Capture weekly architectural and process reviews
- →Organize retrospectives into a time-indexed hierarchy
- →Transform retrospective findings into actionable ledger tasks
- →Retrieve retrospectives within a specific date window
How it works
The skill captures retrospectives at task, daily, and weekly levels using a structured schema including 'Well', 'Not Well', 'Start', 'Stop', 'Continue', and 'Improvements'. It organizes these into a time-indexed hierarchy and can create ledger tasks from findings.
Inputs & outputs
When to use retrospectives
- →Conducting a post-mortem for a failed task
- →Performing a daily session wrap-up
- →Reviewing weekly architectural progress
About this skill
Retrospectives Skill
The Retrospectives skill provides a structured way to capture, organize, and act upon project and session reflections. It transforms raw feedback into actionable ledger tasks to drive iterative improvement.
Core Capabilities
The system organizes retrospectives into a time-indexed hierarchy (Year $\rightarrow$ Month $\rightarrow$ Day) to ensure historical continuity and easy retrieval.
Retrospective Capture
Supports three levels of reflection:
- Task-based: Post-mortems for specific high-impact tasks.
- Daily: Session-end reflections on daily progress.
- Weekly: Higher-level architectural and process reviews.
The Reflection Schema
Every retrospective captures a structured state:
- Well: What went right? (Successes to replicate)
- Not Well: What failed? (Pain points to resolve)
- Start: New behaviors or tools to adopt.
- Stop: Counter-productive patterns to cease.
- Continue: Stable practices to maintain.
- Improvements: Direct actionable goals.
Actionable Ledger
The skill integrates with the project ledger. When a retrospective identifies a critical failure or improvement, it can spawn a ledgerTask, linking the theoretical reflection to an actual work item.
Commands
save_retro
Persists a structured retrospective to the sovereign memory.
- Arguments:
date(YYYY-MM-DD),type(task/day/week),id,content(RetroSchema).
get_retros_for_period
Retrieves all retrospectives within a specific date window.
- Arguments:
start_date,end_date.
create_ledger_task
Transforms a retrospective finding into a formal ledger task.
- Arguments:
description,priority,source_retro_id.
Implementation Details
- Architecture: Two-tier Python implementation (FastMCP $\rightarrow$ RetrospectivesLogic).
- Storage: Filesystem-based hierarchy using JSON documents.
- Organization: Time-series layout for optimized retrieval and historical auditing.
When not to use it
- →When a structured reflection process is not desired
- →When the project ledger is not used for task management
Limitations
- →Retrospectives are organized into a time-indexed hierarchy (Year → Month → Day).
- →The reflection schema includes specific fields: Well, Not Well, Start, Stop, Continue, Improvements.
- →The skill integrates with the project ledger to spawn ledgerTasks.
How it compares
This skill integrates directly with a project ledger to convert retrospective findings into formal tasks, unlike simply documenting reflections without a direct action mechanism.
Compared to similar skills
retrospectives side by side with the closest alternatives in the catalog.
| Skill | Installs | Updated | Safety | Difficulty |
|---|---|---|---|---|
| retrospectives (this skill) | 0 | 2mo | Review | Beginner |
| notion-spec-to-implementation | 5 | 6mo | No flags | Intermediate |
| weekly | 2 | 5mo | No flags | Beginner |
| goal-tracking | 1 | 5mo | No flags | Intermediate |
Try saying
Example prompts that trigger this skill in your AI assistant.
You might also like
notion-spec-to-implementation
openai
Turn Notion specs into implementation plans, tasks, and progress tracking; use when implementing PRDs/feature specs and creating Notion plans + tasks from them.
weekly
ballred
Facilitate weekly review process with reflection, goal alignment, and planning. Create review notes, analyze past week, plan next week. Use on Sundays or whenever doing weekly planning.
goal-tracking
ballred
Track progress toward 3-year, yearly, monthly, and weekly goals. Calculate completion percentages, surface stalled goals, connect daily tasks to objectives. Use for goal reviews and progress tracking.
master-plan-manager
ananddtyagi
Safe MASTER_PLAN.md management with backup, validation, and intelligent updates. Use when updating task tracking, adding features to roadmap, or modifying project documentation.
decision-logger
alirezarezvani
Two-layer memory architecture for board meeting decisions. Manages raw transcripts (Layer 1) and approved decisions (Layer 2). Use when logging decisions after a board meeting, reviewing past decisions with /cs:decisions, or checking overdue action items with /cs:review. Invoked automatically by the board-meeting skill after Phase 5 founder approval.
gsd-plant-seed
phorde
Capture a forward-looking idea with trigger conditions — surfaces automatically at the right milestone