Agent skill

Retrospective

Analyze the current session for improvement opportunities in skills, memory, and workflow.

MITAuto-check: notesProduct & Project Management

Install Retrospective

skills CLI

$ npx skills add Szotasz/marveen --skill retrospective -a claude-code

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

GitHub CLI

$ gh skill install Szotasz/marveen retrospective --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/Szotasz/marveen.git skills-src && mkdir -p .claude/skills && cp -r skills-src/seed-skills/retrospective .claude/skills/retrospective && 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/

Paste into Claude Code

Install the "retrospective" agent skill from https://github.com/Szotasz/marveen/tree/develop/seed-skills/retrospective into .claude/skills/retrospective/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "retrospective", then confirm the skill loads.

Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

$skill-installer

$skill-installer install https://github.com/Szotasz/marveen/tree/develop/seed-skills/retrospective

Type this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.

skills CLI

$ npx skills add Szotasz/marveen --skill retrospective -a codex

Project install goes to .agents/skills/; add -g for ~/.codex/skills/.

GitHub CLI

$ gh skill install Szotasz/marveen retrospective --agent codex

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/Szotasz/marveen.git skills-src && mkdir -p .agents/skills && cp -r skills-src/seed-skills/retrospective .agents/skills/retrospective && rm -rf skills-src

Use ~/.agents/skills/ instead of .agents/skills for a personal install.

Codex skills documentation · loads skills from .agents/skills/

Paste into Codex

Install the "retrospective" agent skill from https://github.com/Szotasz/marveen/tree/develop/seed-skills/retrospective into .agents/skills/retrospective/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "retrospective", then confirm the skill loads.

Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

skills CLI

$ npx skills add Szotasz/marveen --skill retrospective -a cursor

Project install goes to .agents/skills/; add -g for ~/.cursor/skills/.

GitHub CLI

$ gh skill install Szotasz/marveen retrospective --agent cursor

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/Szotasz/marveen.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/seed-skills/retrospective .cursor/skills/retrospective && rm -rf skills-src

Use ~/.cursor/skills/ instead of .cursor/skills for a personal install.

Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/

Paste into Cursor

Install the "retrospective" agent skill from https://github.com/Szotasz/marveen/tree/develop/seed-skills/retrospective into .cursor/skills/retrospective/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "retrospective", then confirm the skill loads.

Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

gemini skills install

$ gemini skills install https://github.com/Szotasz/marveen.git --path seed-skills/retrospective

--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.

skills CLI

$ npx skills add Szotasz/marveen --skill retrospective -a gemini-cli

Project install goes to .agents/skills/; add -g for ~/.gemini/skills/.

GitHub CLI

$ gh skill install Szotasz/marveen retrospective --agent gemini-cli

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/Szotasz/marveen.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/seed-skills/retrospective .gemini/skills/retrospective && rm -rf skills-src

Use ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.

Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/

Paste into Gemini CLI

Install the "retrospective" agent skill from https://github.com/Szotasz/marveen/tree/develop/seed-skills/retrospective into .gemini/skills/retrospective/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "retrospective", then confirm the skill loads.

Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

GitHub CLI

$ gh skill install Szotasz/marveen retrospective

Installs for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).

skills CLI

$ npx skills add Szotasz/marveen --skill retrospective -a github-copilot

Project install goes to .agents/skills/; add -g for ~/.copilot/skills/.

Manual copy

$ git clone --depth 1 https://github.com/Szotasz/marveen.git skills-src && mkdir -p .github/skills && cp -r skills-src/seed-skills/retrospective .github/skills/retrospective && rm -rf skills-src

Use ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.

GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/

Paste into GitHub Copilot

Install the "retrospective" agent skill from https://github.com/Szotasz/marveen/tree/develop/seed-skills/retrospective into .github/skills/retrospective/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "retrospective", then confirm the skill loads.

GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

skills CLI

$ npx skills add Szotasz/marveen --skill retrospective -a opencode

OpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.

GitHub CLI

$ gh skill install Szotasz/marveen retrospective --agent opencode

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/Szotasz/marveen.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/seed-skills/retrospective .opencode/skills/retrospective && rm -rf skills-src

Use ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.

OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/

Paste into OpenCode

Install the "retrospective" agent skill from https://github.com/Szotasz/marveen/tree/develop/seed-skills/retrospective into .opencode/skills/retrospective/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "retrospective", then confirm the skill loads.

OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

Facts

Skill name: retrospective
GitHub stars: 117
Token cost: ~1.5k tokens
SKILL.md length: 469 words
Files: 1
Skills in repo: 18
Repo updated: Oct 10, 2026
First seen: Oct 7, 2026
Licence: MIT

At a glance

Analyze the current session for improvement opportunities in skills, memory, and workflow.

Works in 5 steps: Gather session data → Analyze with A/B/C framework → Generate improvement proposals → …
A session involved complex problem-solving
SKILL.md covers When to use, Arguments, Procedure and Pitfalls, plus 1 more section
Calls curl

What it does

Retrospective is an agent skill from Szotasz/marveen. Analyze the current session for improvement opportunities in skills, memory, and workflow. Spawns a sub-agent for unbiased analysis. Use when a session involved complex problem-solving, error recovery, user corrections, or multi-step workflows. Trigger on "/retrospective" command or at session end after significant work.

Its SKILL.md is about 1.5k 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, covering Retrospectives and Subagents. The repository describes itself as: AI csapatod, ami fut amíg te alszol. The licence is MIT.

When your agent uses it

A session involved complex problem-solving
User corrections
Multi-step workflows
/retrospective command

Example prompts

“/retrospective”
“/retrospective”

Workflow steps

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

Gather session data
Analyze with A/B/C framework
Generate improvement proposals
Present to user for approval
Execute approved changes

What it can do on your machine

Read from SKILL.md and the folder at commit bcd90b2. 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
Shell commands in SKILL.md call:
- curl
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md. Its commands use curl, which can reach the network depending on how they are called.
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

Retrospective loads about 1.5k tokens when it runs. Until then it costs about 84 tokens; SKILL.md has 469 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it: ~84
When it runs · the whole SKILL.md, loaded when a task matches: ~1.5k

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: notes

The automated check noted patterns worth knowing about, such as sudo or a known installer.

NoteMentions a .env fileSKILL.md:44

PORT="$(sed -n 's/^WEB_PORT=//p' .env 2>/dev/null | head -1 | tr -d '"')"; PORT="${PORT:-3420}"

NoteMentions a .env fileSKILL.md:149

PORT="$(sed -n 's/^WEB_PORT=//p' .env 2>/dev/null | head -1 | tr -d '"')"; PORT="${PORT:-3420}"

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 Szotasz/marveen at commit bcd90b2, republished under its MIT licence (© Szotasz). 469 words, ~1,534 tokens.

Download SKILL.mdSave it as .claude/skills/retrospective/SKILL.md (or your agent's skills folder).

name: retrospective
description: Analyze the current session for improvement opportunities in skills, memory, and workflow. Spawns a sub-agent for unbiased analysis. Use when a session involved complex problem-solving, error recovery, user corrections, or multi-step workflows. Trigger on "/retrospective" command or at session end after significant work.

Retrospective -- Session Analysis & Improvement

When to use

After a complex session (5+ tool calls, error recovery, multi-step workflow)
When the user corrected your approach ("no, do it this way")
After a failed attempt that required a different strategy
Before context window exhaustion on a productive session
Explicitly via /retrospective command

Arguments

Argument	Required	Description
`scope`	no	What to analyze: `session` (default), `last-task`, `last-hour`
`focus`	no	Narrow analysis: `skills`, `memory`, `workflow`, or `all` (default)

Examples:

/retrospective -- full session analysis
/retrospective scope=last-task focus=skills -- only skill improvements for the last task
/retrospective focus=memory -- only memory tier/content suggestions

Procedure

1. Gather session data

Collect from the current conversation context:

What tasks were attempted
Which approaches succeeded vs failed
User corrections and their reasoning
Tools/skills used and their effectiveness
Errors encountered and how they were resolved
Time spent on dead ends vs productive work

Also pull external state:

bash

AGENT_ID="$(echo $BOT_NAME | tr '[:upper:]' '[:lower:]')"
PORT="$(sed -n 's/^WEB_PORT=//p' .env 2>/dev/null | head -1 | tr -d '"')"; PORT="${PORT:-3420}"

# Recent memories written this session
curl -s -H "Authorization: Bearer $(cat store/.dashboard-token)" \
  "http://localhost:$PORT/api/memories?agent=$AGENT_ID&category=hot&limit=20"

# Today's daily log entries
DATE=$(date +%Y-%m-%d)
curl -s -H "Authorization: Bearer $(cat store/.dashboard-token)" \
  "http://localhost:$PORT/api/daily-log?agent=$AGENT_ID&date=$DATE"

# Skills that were referenced or used
ls ~/.claude/skills/ | head -30

2. Analyze with A/B/C framework

For each significant event in the session, evaluate:

A -- What happened? State the fact: what was attempted, what was the outcome.

B -- Why did it happen that way? Root cause: was it a missing skill, wrong memory tier, incorrect assumption, tooling gap, or communication issue?

C -- What should change? Concrete action: new skill, skill patch, memory write/update, workflow change, or nothing (if the outcome was correct).

3. Generate improvement proposals

Organize findings into categories:

Skill proposals

For each skill-related finding:

SKILL_ACTION: create | patch | delete
SKILL_NAME: name-of-skill
REASON: why this change helps
CHANGE: what specifically to add/modify/remove

Rules:

Only propose a NEW skill if the pattern appeared 2+ times or was complex enough (5+ steps)
Prefer PATCH over CREATE -- check if an existing skill covers 80% of the case
Check ~/.claude/skills/.skill-index.md before proposing duplicates

Memory proposals

For each memory-related finding:

MEMORY_ACTION: save | update | delete | retier
MEMORY_TIER: hot | warm | cold | shared
CONTENT: what to remember
REASON: why this tier, why now

Rules:

User corrections -> always save as warm (stable preference)
Task-specific findings -> hot (will decay naturally)
Architectural decisions -> cold (long-term reference)
Cross-agent learnings -> shared

Show full SKILL.md (175 more words)Show less

Workflow proposals

For process improvements that don't fit skills or memory:

WORKFLOW_CHANGE: description of the process change
APPLIES_TO: this agent | all agents | specific agent
REASON: what problem it solves

4. Present to user for approval

Format the output as a concise action list:

## Retrospective Summary

Session: [brief description]
Duration: ~[estimate]
Key events: [count] tasks, [count] corrections, [count] errors

### Proposed Changes

#### Skills
1. [PATCH] skill-name: add X because Y
2. [CREATE] new-skill: handles Z pattern (seen 3x this session)

#### Memory
1. [SAVE warm] "user prefers X over Y" -- correction at [context]
2. [RETIER hot->cold] "project Z deadline" -- no longer active

#### Workflow
1. [ALL AGENTS] Always check CI before merging -- 2 failed merges this session

Apply all? (y/n/select)

5. Execute approved changes

On user approval:

Skills: create/patch SKILL.md files, regenerate .skill-index.md
Memory: write/update via the memory API
Workflow: update CLAUDE.md or inter-agent message to affected agents

After execution, log the retrospective to the daily log:

bash

PORT="$(sed -n 's/^WEB_PORT=//p' .env 2>/dev/null | head -1 | tr -d '"')"; PORT="${PORT:-3420}"
curl -s -X POST -H "Authorization: Bearer $(cat store/.dashboard-token)" \
  http://localhost:$PORT/api/daily-log \
  -H "Content-Type: application/json" \
  -d "{\"agent_id\":\"$AGENT_ID\",\"content\":\"## $(date +%H:%M) -- Retrospective\n[count] skill changes, [count] memory updates, [count] workflow changes applied.\"}"

Pitfalls

Do NOT auto-apply changes without user approval -- always present first
Do NOT create skills for one-off tasks that won't repeat
Do NOT save ephemeral debugging context as cold memory
If the session was straightforward with no issues, say so and skip -- not every session needs changes
Keep proposals actionable and specific -- "be better at X" is not a proposal

Relation to existing mechanisms

Mechanism	Retrospective's role
memoria-heartbeat	Retrospective is the dedicated, deeper version. Heartbeat does inline A/B/C; retrospective does it thoroughly with user approval
skill-factory	Retrospective proposes skills; skill-factory creates them. Retrospective may invoke skill-factory for CREATE actions
DREAM.md	Nightly consolidation. Retrospective is on-demand, immediate
/handoff	Retrospective analyzes; handoff transfers. Different purposes

Files

Just SKILL.md in seed-skills/retrospective of Szotasz/marveen.

Open the folder on GitHubat commit bcd90b2

Compare with similar skills

Retrospective 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.

Retrospective compared with similar skills
Skill	Stars	Used in	Tokens	Auto-check	Licence	Repo updated
Retrospective this skillSzotasz/marveen	117	—	~1.5k	Automated safety check: Notes	MIT	today
Skill Retrokenneth-liao/ai-launchpad-marketplace	126	—	~2.1k	Automated safety check: Pass	None	6 mo ago
Oracle PrismSoul-Brews-Studio/arra-oracle-skills-cli	123	—	~1.3k	Automated safety check: Pass	MIT	8 days ago
Visualization Ops SkillOpenLoaf/OpenLoaf	108	—	~2.5k	Automated safety check: Pass	AGPL-3.0	4 mo ago
Set Workflow Statesandgardenhq/sgai	137	—	~1.2k	Automated safety check: Pass	Custom licence	20 days ago
Processnotque/vexjoy-agent	441	—	~2.1k	Automated safety check: Notes	MIT	yesterday

Similar skills

Skill Retro
kenneth-liao/ai-launchpad-marketplace
A skill your agent uses when reviewing how skills performed during a session, when the user wants to analyze skill invocations and identify improvements, or when the user says "skill retro", "review…
126 GitHub stars~2.1k tokensUpdated 6 mo ago
Product & Project ManagementAuto-check passed
Oracle Prism
Soul-Brews-Studio/arra-oracle-skills-cli
Multi-perspective analysis — one agent transforms through 5 lenses (Archaeologist, Bug Hunter, Skeptic, Architect, Auditor) inline, no subagents.
123 GitHub stars~1.3k tokensUpdated 8 days ago
Product & Project ManagementAuto-check passed
Visualization Ops Skill
OpenLoaf/OpenLoaf
Required for visualized output. An agent skill from OpenLoaf/OpenLoaf.
108 GitHub stars~2.5k tokensUpdated 4 mo ago
Product & Project ManagementAuto-check passed
Set Workflow State
sandgardenhq/sgai
Coordinator and retrospective workflow-state protocol. An agent skill from sandgardenhq/sgai.
137 GitHub stars~1.2k tokensUpdated 20 days ago
Product & Project ManagementAuto-check passed
Process
notque/vexjoy-agent
Process: retrospectives, session handoff, pair programming, subagent-driven development, condition-based waiting.
441 GitHub stars~2.1k tokensUpdated yesterday
Agent WorkflowsAuto-check: notes
Bad
stephenleo/bmad-autonomous-development
BMad Autonomous Development — orchestrates parallel story implementation pipelines.
107 GitHub stars~7.7k tokensUpdated 5 mo ago
Agent WorkflowsAuto-check passed

Questions about Retrospective

What does Retrospective do?

Analyze the current session for improvement opportunities in skills, memory, and workflow. Retrospective is an agent skill from Szotasz/marveen. Analyze the current session for improvement opportunities in skills, memory, and workflow.

When should I use Retrospective?

Retrospective fits situations like: A session involved complex problem-solving; user corrections; multi-step workflows; /retrospective command.

How do I install Retrospective in Claude Code?

Run `npx skills add Szotasz/marveen --skill retrospective -a claude-code`. Or copy the skill folder (seed-skills/retrospective in Szotasz/marveen) into .claude/skills/retrospective in your project. Claude Code loads it when a task matches its description.

How do I install Retrospective in Codex?

Run `npx skills add Szotasz/marveen --skill retrospective -a codex`. Or copy the skill folder (seed-skills/retrospective in Szotasz/marveen) into .agents/skills/retrospective in your project. Codex loads it when a task matches its description.

Can I use Retrospective 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 Szotasz/marveen --skill retrospective -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/retrospective, .gemini/skills/retrospective, .github/skills/retrospective and .opencode/skills/retrospective in your project.

What does Retrospective need to run?

Going by SKILL.md and its folder, Retrospective needs the command-line tools its instructions call (curl).

Does Retrospective access the network?

SKILL.md contains no URLs. Its commands use curl, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Retrospective safe to install?

Our automated static check of SKILL.md found notes only (mentions a .env file), nothing it rates as a warning. It is not a guarantee. Review the folder before installing.

What licence does Retrospective use?

Retrospective 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 Retrospective use?

About 1.5k tokens (SKILL.md is roughly 6.1k 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 Retrospective?

Skills that share tags, products or a category with Retrospective: Skill Retro (kenneth-liao/ai-launchpad-marketplace, 126 stars), Oracle Prism (Soul-Brews-Studio/arra-oracle-skills-cli, 123 stars), Visualization Ops Skill (OpenLoaf/OpenLoaf, 108 stars) and Set Workflow State (sandgardenhq/sgai, 137 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Retrospective?

Szotasz (a GitHub user) maintains it in Szotasz/marveen, which has 117 GitHub stars. The repository holds 18 skills in this directory. The repository was last updated on October 10, 2026.

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

Install Retrospective

Facts

At a glance

What it does

When your agent uses it

Example prompts

Workflow steps

What it can do on your machine

Tool permissions

Runs code

Network

Credentials

Context cost

Safety

SKILL.md

Retrospective -- Session Analysis & Improvement

When to use

Arguments

Procedure

1. Gather session data

2. Analyze with A/B/C framework

3. Generate improvement proposals

Skill proposals

Memory proposals

Workflow proposals

4. Present to user for approval

5. Execute approved changes

Pitfalls

Relation to existing mechanisms

Files

Compare with similar skills

Similar skills

More from Szotasz/marveen

Related

Questions about Retrospective