Agent skill

Retro

by MAGA2010 in MAGA2010/hackathon-run

Reviews the hackathon journey after submission and produces a prioritized list of learnings for next time.

MITAuto-check passedProduct & Project Management

Install Retro

skills CLI
$ npx skills add MAGA2010/hackathon-run --skill retro -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install MAGA2010/hackathon-run retro --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/MAGA2010/hackathon-run.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/retro .claude/skills/retro && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
retro
GitHub stars
367
Token cost
~1k tokens
SKILL.md length
332 words
Files
1
Skills in repo
15
Repo updated
First seen
Licence
MIT

At a glance

Reviews the hackathon journey after submission and produces a prioritized list of learnings for next time.

  • Works in 4 steps: Compute the four ratios → Mine the surprises → Generate the action list → …
  • Product & Project Management work in your project
  • SKILL.md covers Input contract, Execution, Output contract and Acceptance criteria, plus 2 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Retro is an agent skill from MAGA2010/hackathon-run. Reviews the hackathon journey after submission and produces a prioritized list of learnings for next time. Use after ship-pack has marked the project ready to submit; do not invoke mid-build.

Its SKILL.md is about 1k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Product & Project Management. The licence is MIT.

When your agent uses it

  • Product & Project Management work in your project

Example prompts

  • “Use the retro skill to review the hackathon journey after submission and produces a prioritized list of learnings for next time”
  • “/retro”

Workflow steps

4 steps, taken from the step headings in SKILL.md.

  1. Compute the four ratios
  2. Mine the surprises
  3. Generate the action list
  4. Emit the one-page PDF-ready summary

What it can do on your machine

Read from SKILL.md and the folder at commit de3bcfb. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Retro loads about 1k tokens when it runs. Until then it costs about 49 tokens; SKILL.md has 332 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~49
When it runs · the whole SKILL.md, loaded when a task matches
~1k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from MAGA2010/hackathon-run at commit de3bcfb, republished under its MIT licence (© MAGA2010). 332 words, ~1,045 tokens.

Download SKILL.mdSave it as .claude/skills/retro/SKILL.md (or your agent's skills folder).
name
retro
description
Reviews the hackathon journey after submission and produces a prioritized list of learnings for next time. Use after ship-pack has marked the project ready to submit; do not invoke mid-build.
when_to_use
Trigger when the user says "what did we learn", "post-mortem", "retrospective", "what should we do differently next time", or after ship-pack marks all gates…
version
1.0
category
lifecycle
tags
["post-mortem", "ratios", "learn"]
dependencies
["decision-log", "ship-pack"]
side_effects
["retro"]
triggers
["retrospective", "what did we learn", "post-mortem", "keep doing stop doing", "try next time"]

retro

The hackathon does not end at submission. The team has roughly an hour of high-fidelity memory before it decays. retro captures that memory in a form a future team can actually use.

Input contract

Required:

  • .hackathon/state/plan.json (load; the KEEP / CUT / DEFER decisions)
  • .hackathon/state/verify.json (load; what actually worked)
  • .hackathon/state/review.json (load; judge scores)
  • .hackathon/state/ship.json (load; submission outcome)
  • .hackathon/state/time-box.json (load; clock usage vs plan)

Optional:

  • free_text_notes: anything the team said in the last hour
  • next_event_date: ISO 8601 string; used for follow-up reminders

Execution

1. Compute the four ratios
RatioFormula
scope_accuracyKEEP / (KEEP + CUT + DEFER)
time_accuracyplanned_minutes / actual_minutes
verify_pass_ratepassed_steps / total_steps
judge_score_avgmean(score across all judges)

Print each with a one-sentence interpretation.

2. Mine the surprises

Compare plan vs reality and emit a surprises list:

  • A feature that was CUT but turned out to be trivial
  • A feature that was KEEP but took > 2x the time
  • A failure during verify that was foreseeable
  • A judge dimension that scored lower than expected
3. Generate the action list

Three buckets, each with at most 3 items:

  • keep_doing: behaviors to repeat
  • stop_doing: behaviors to drop
  • try_next_time: experiments for the next event

Each item has: behavior (one sentence), evidence (which state file shows it), owner_hint (which role would own it next time).

4. Emit the one-page PDF-ready summary
text
# Retro — <event name> — <date>

## Four ratios
- scope_accuracy: 0.42 (cut more than half; mild over-scoping)
- time_accuracy:  1.15 (slight overrun; buffer held)
- verify_pass_rate: 0.83 (one flaky step near demo)
- judge_score_avg: 3.7 / 5 (mid-pack)

## Surprises
- Login was CUT but trivial in retrospect.

## Keep doing
- Pair-programmed the demo path (evidence: verify.json).

## Stop doing
- Spent 2h on a polished README (evidence: ship.json).

## Try next time
- Schedule a dry-run 2h before the live demo.

Output contract

Files written:

  • .hackathon/state/retro.json (NEW schema)
  • .hackathon/artifacts/retro.md (the human-readable summary)

Acceptance criteria

  • All four ratios are computed and printed.
  • Surprises list has at least one item OR is explicitly empty with rationale.
  • Each bucket (keep_doing, stop_doing, try_next_time) has <= 3 items.
  • Every action item has an evidence pointer to a state file.
  • retro.md is <= 250 lines and stands alone as a post-mortem.

Failure modes

ModeBehavior
Missing state filesSkip the corresponding ratio; note it
judge_score_avg = NaNSubstitute not judged placeholder
free_text_notes emptySkip the free-text section
Ran during buildRefuse; suggest running after ship-pack

Trigger phrases (for agent intent matching)

  • "what did we learn"
  • "post-mortem"
  • "retrospective"
  • "what should we do differently next time"
  • "wrap up the hackathon"
  • "lessons learned"

© MAGA2010, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in skills/retro of MAGA2010/hackathon-run.

Open the folder on GitHubat commit de3bcfb

Compare with similar skills

Retro next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Retro compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Retro this skillMAGA2010/hackathon-run367—~1kAutomated safety check: PassMIT
User Story Writerdeanpeters/Product-Manager-Skills7.2k2 repos~2.9kAutomated safety check: PassCustom licence
Game Changing FeaturesopenstatusHQ/data-table-filters2.3k3 repos~2.1kAutomated safety check: PassMIT
CCPM Project Managementautomazeio/ccpm8.4k—~1.1kAutomated safety check: PassMIT
Convex Create Componentspokvulcan/poker-planning1148 repos~2.6kAutomated safety check: PassMIT
Self Improving Agentfarm-fe/farm5.6k2 repos~3.3kAutomated safety check: NotesMIT

Similar skills

  • User Story Writer

    deanpeters/Product-Manager-Skills

    Writes user stories in Mike Cohn's format with Gherkin acceptance criteria, turning user needs into development-ready work with testable conditions.

    7.2k GitHub starsUsed in 2 repos~2.9k tokens
    Product & Project ManagementAuto-check passed
  • Game Changing Features

    openstatusHQ/data-table-filters

    Find 10x product opportunities and high-leverage improvements.

    2.3k GitHub starsUsed in 3 repos~2.1k tokens
    Product & Project ManagementAuto-check passed
  • Runs a spec-driven workflow from PRD to epic to GitHub issues to parallel agents, with status, standup and blocked-work reports from bundled scripts.

    8.4k GitHub stars~1.1k tokensUpdated 6 mo ago
    Product & Project ManagementAuto-check passed
  • Convex Create Component

    spokvulcan/poker-planning

    Builds reusable Convex components with isolated tables and app-facing APIs.

    114 GitHub starsUsed in 8 repos~2.6k tokens
    Product & Project ManagementAuto-check passed
  • A universal self-improving agent that learns from ALL skill experiences.

    5.6k GitHub starsUsed in 2 repos~3.3k tokens
    Product & Project ManagementAuto-check: notes
  • Builds a weekly engineering retrospective from git history: commit counts, per-person contributions, work patterns and code quality numbers over a chosen window.

    136k GitHub stars~2.4k tokensUpdated yesterday
    Product & Project ManagementAuto-check passed

More from MAGA2010/hackathon-run

All 15 skills in this repo
  • Decision Log

    MAGA2010/hackathon-run

    Writes every KEEP/CUT/DEFER/PIVOT scope decision into an append-only team log with rationale, author, and timestamp.

    367 GitHub stars~847 tokensUpdated yesterday
    Auto-check passed
  • Demo Coach

    MAGA2010/hackathon-run

    Drafts a 30, 60, or 90-second pitch script for the demo, structured as opening, pain, product, action, result, close.

    367 GitHub stars~895 tokensUpdated yesterday
    Auto-check passed
  • Judge Sim

    MAGA2010/hackathon-run

    Scores a hackathon project 0 to 5 across seven judging dimensions and produces a prioritized fix list for the last hour.

    367 GitHub stars~970 tokensUpdated yesterday
    Auto-check passed
  • Scope Knife

    MAGA2010/hackathon-run

    Forces a KEEP, CUT, or DEFER decision on every feature when scope is too large, no MVP consensus exists, or time is running out.

    367 GitHub stars~1.3k tokensUpdated yesterday
    Auto-check passed
  • Time Box

    MAGA2010/hackathon-run

    Allocates time across the hackathon lifecycle (idea - scope - build - verify - demo - ship) and warns before each deadline slips.

    367 GitHub stars~1.3k tokensUpdated yesterday
    Auto-check passed
  • Fast Verify

    MAGA2010/hackathon-run

    Verifies the demo path runs end-to-end by executing each step in order, recording the actual outcome, and stopping at the first failure with a diagnostic.

    367 GitHub stars~893 tokensUpdated yesterday
    Auto-check passed

Questions about Retro

What does Retro do?

Reviews the hackathon journey after submission and produces a prioritized list of learnings for next time. Retro is an agent skill from MAGA2010/hackathon-run. Reviews the hackathon journey after submission and produces a prioritized list of learnings for next time.

When should I use Retro?

Retro fits situations like: product & Project Management work in your project.

How do I install Retro in Claude Code?

Run `npx skills add MAGA2010/hackathon-run --skill retro -a claude-code`. Or copy the skill folder (skills/retro in MAGA2010/hackathon-run) into .claude/skills/retro in your project. Claude Code loads it when a task matches its description.

How do I install Retro in Codex?

Run `npx skills add MAGA2010/hackathon-run --skill retro -a codex`. Or copy the skill folder (skills/retro in MAGA2010/hackathon-run) into .agents/skills/retro in your project. Codex loads it when a task matches its description.

Can I use Retro in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add MAGA2010/hackathon-run --skill retro -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/retro, .gemini/skills/retro, .github/skills/retro and .opencode/skills/retro in your project.

What does Retro need to run?

SKILL.md names no scripts, command-line tools or credentials: Retro is instructions for the agent only.

Does Retro access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Retro safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Retro use?

Retro is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Retro use?

About 1k tokens (SKILL.md is roughly 4.2k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Retro?

Skills that share tags, products or a category with Retro: User Story Writer (deanpeters/Product-Manager-Skills, 7.2k stars), Game Changing Features (openstatusHQ/data-table-filters, 2.3k stars), CCPM Project Management (automazeio/ccpm, 8.4k stars) and Convex Create Component (spokvulcan/poker-planning, 114 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Retro?

MAGA2010 (a GitHub user) maintains it in MAGA2010/hackathon-run, which has 367 GitHub stars. The repository holds 15 skills in this directory. The repository was last updated on October 7, 2026.

Source: MAGA2010/hackathon-run on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.