# Create Learnings

> Capture learnings after completing a feature, sprint, or project in a structured retrospective format.

- **Type:** Skill
- **Install:** `agentstack add skill-tomzx-agents-create-learnings`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [tomzx](https://agentstack.voostack.com/s/tomzx)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [tomzx](https://github.com/tomzx)
- **Source:** https://github.com/tomzx/agents/tree/main/skills/create-learnings

## Install

```sh
agentstack add skill-tomzx-agents-create-learnings
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Create Learnings

Facilitates a retrospective to capture actionable learnings after completing a feature, sprint, or project.
Produces a structured document covering what went well, what didn't, process improvements, technical insights, and next actions.

## Prerequisites

- Apply the shared SDLC conventions in `skills/sdlc/references/shared.md`.
- If no argument is provided, locate the feature directory under `.sdlc/features/` whose frontmatter `issue` field references `$ISSUE_NUMBER`.
- A completed feature, sprint, or project to reflect on
- Context about what was built, how long it took, and any notable events
- If any files exist under `.sdlc/knowledge/assumptions/` or `.sdlc/knowledge/decisions/` for this feature, review them for context.

## Steps

1. Gather context: what was delivered, timeline, team involved, and any notable events.
2. Reflect on what went well (practices worth repeating and amplifying).
3. Reflect on what didn't go well, identifying root causes not just symptoms.
4. Identify concrete process improvements with owners and dates.
5. Capture technical insights: decisions that paid off and decisions to revisit.
6. Distill actionable next steps.
7. Write the output to `.sdlc/knowledge/learnings/N-.md` where N is the next available sequence number in that directory.

## Output Format

Use the template at `skills/sdlc/templates/knowledge/learning.md` (copied to `.sdlc/templates/knowledge/learning.md` by `/initialize-sdlc-directory`; use the project's customized copy if present). Write the result to the artifact path named in the steps above.

## Outcome

If `$OUTCOME_YAML` is set, emit `verdict: approved` there per `skills/sdlc/references/shared.md` once the learnings artifact is written.

## Example Usage

**Scenario 1: Feature retrospective**
A payment feature took 3 weeks instead of 2.
Learnings: the third-party API was underdocumented (add a spike phase to future plans involving new integrations), automated integration tests caught 4 regressions early (keep and expand), the spec was changed mid-implementation (add a spec-freeze milestone to the plan template).

**Scenario 2: Sprint retrospective**
End-of-sprint review with the team.
What went well: daily standups kept everyone aligned.
What didn't: unclear task definitions led to rework.
Action item: run `/create-tasks-decomposition` before each sprint starts.

## Completion Checklist

Before handing off to review, confirm:

- [ ] "What could improve" entries identify root causes, not just symptoms
- [ ] Action items are specific, owned, and dated

Self-check the draft against the [`review-learnings` checklist](../review-learnings/SKILL.md) and fix what you can, so review finds less to flag.

## Useful Commands Reference

No CLI commands required. This skill operates on information provided in context.

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [tomzx](https://github.com/tomzx)
- **Source:** [tomzx/agents](https://github.com/tomzx/agents)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-tomzx-agents-create-learnings
- Seller: https://agentstack.voostack.com/s/tomzx
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.