Skip to content
asciinema-recorder logo

Asciinema Recorder

asciinema-recorder

Record Claude Code sessions with asciinema. TRIGGERS - record session, asciinema record, capture terminal, demo recording.

terrylica/cc-skills0installs75stars

SKILL.md

Full skill instructions

asciinema-recorder

Generate ready-to-copy commands for recording Claude Code sessions with asciinema. Dynamically creates filenames based on workspace and datetime.

Platform: macOS, Linux (requires asciinema CLI)

Self-Evolving Skill: This skill improves through use. If instructions are wrong, parameters drifted, or a workaround was needed — fix this file immediately, don't defer. Only update for real, reproducible issues.

When to Use This Skill

Use this skill when:

  • Starting a new terminal recording session
  • Creating demos or documentation recordings
  • Capturing Claude Code sessions for review or sharing
  • Generating workspace-specific recording filenames

Why This Skill?

This skill generates ready-to-copy recording commands with:

  • Dynamic workspace-based filename
  • Datetime stamp for uniqueness
  • Saves to project's tmp/ folder (gitignored)

Requirements

ComponentRequiredInstallation
asciinema CLIYesbrew install asciinema

Workflow Phases

Phase 0: Preflight Check

Purpose: Verify asciinema is installed.

# Check asciinema CLI
which asciinema && asciinema --version

If asciinema is NOT installed, use AskUserQuestion:

  • question: "asciinema not found. How would you like to proceed?" header: "Setup" multiSelect: false options:
    • label: "Install asciinema (Recommended)" description: "Run: brew install asciinema (macOS) or apt install asciinema (Linux)"
    • label: "Show manual instructions" description: "Display installation commands for all platforms"
    • label: "Cancel" description: "Exit without recording"

Based on selection:

  • "Install asciinema" → Run appropriate install command based on OS:

    # macOS
    brew install asciinema
    
    # Linux (apt)
    sudo apt install asciinema
    
    # Linux (pip)
    pip install asciinema
    

    Then proceed to Phase 1.0.

  • "Show manual instructions" → Display the commands above, then exit skill.

  • "Cancel" → Exit skill without action.


Phase 1.0: Recording Location

Purpose: Let user choose where to save the recording.

Use AskUserQuestion:

  • question: "Where should the recording be saved?" header: "Location" multiSelect: false options:
    • label: "$PWD/​tmp/ (Recommended)" description: "Project tmp directory (gitignored)"
    • label: "~/​asciinema/" description: "Global recordings directory"
    • label: "Custom path" description: "Specify your own directory"

Based on selection:

  • "$PWD/​tmp/" → Set OUTPUT_DIR="$PWD/​tmp"
  • "~/​asciinema/" → Set OUTPUT_DIR="$HOME/​asciinema" and create if missing: mkdir -p ~/​asciinema
  • "Custom path" → Use user's "Other" input as OUTPUT_DIR

Phase 1.1: Recording Options

Purpose: Let user configure recording behavior.

Use AskUserQuestion:

  • question: "Which recording options would you like?" header: "Options" multiSelect: true options:
    • label: "Add title/​description" description: "Include session title in recording metadata (-t flag)"
    • label: "Disable idle time limit" description: "Keep full pauses instead of 2s max (--idle-time-limit 0)"
    • label: "Quiet mode" description: "Suppress asciinema status messages (-q flag)"

Based on selections, build command flags:

  • "Add title" → Continue to title selection question, add -t "title" flag
  • "Disable idle time limit" → Add --idle-time-limit 0 flag
  • "Quiet mode" → Add -q flag

If "Add title" was selected, follow up with:

  • question: "Enter a title for this recording:" header: "Title" multiSelect: false options:
    • label: "Use workspace name" description: "Title: ${WORKSPACE}"
    • label: "Use workspace + datetime" description: "Title: ${WORKSPACE} ${DATETIME}"
    • label: "Custom title" description: "Enter your own title"

Based on title selection:

  • "Use workspace name" → Set TITLE="${WORKSPACE}"
  • "Use workspace + datetime" → Set TITLE="${WORKSPACE} ${DATETIME}"
  • "Custom title" → Use user's "Other" input as TITLE

Phase 1.2: Detect Context & Generate Command

Purpose: Generate a copy-paste ready recording command.

Step 1.2.1: Detect Workspace

Extract workspace name from $PWD:

/​usr/​bin/​env bash << 'SKILL_SCRIPT_EOF'
WORKSPACE=$(basename "$PWD")
echo "Workspace: $WORKSPACE"
SKILL_SCRIPT_EOF
Step 1.2.2: Generate Datetime
/​usr/​bin/​env bash << 'SKILL_SCRIPT_EOF_2'
DATETIME=$(date +%Y-%m-%d_%H-%M)
echo "Datetime: $DATETIME"
SKILL_SCRIPT_EOF_2
Step 1.2.3: Ensure Output Directory Exists

Create the output directory selected in Phase 1.0:

mkdir -p "${OUTPUT_DIR}"
Step 1.2.4: Construct Command

Build the recording command using:

  • OUTPUT_DIR from Phase 1.0 (location selection)
  • Flags from Phase 1.1 (options selection)
  • TITLE if "Add title" was selected
# Base command
CMD="asciinema rec"

# Add flags from Phase 1.1 options
# (Claude builds this based on user selections)

# Final command format:
asciinema rec ${FLAGS} "${OUTPUT_DIR}/​${WORKSPACE}_${DATETIME}.cast"

Example outputs:

# Default (no options selected):
asciinema rec /​home/​user/​projects/​my-app/​tmp/​my-app_2025-12-21_14-30.cast

# With title + quiet mode:
asciinema rec -t "my-app Demo" -q /​home/​user/​projects/​my-app/​tmp/​my-app_2025-12-21_14-30.cast

# With all options:
asciinema rec -t "my-app 2025-12-21 14:30" -q --idle-time-limit 0 ~/​asciinema/​my-app_2025-12-21_14-30.cast

Phase 2: User Guidance

Purpose: Explain the recording workflow step-by-step.

Present these instructions:

## Recording Instructions

1. **Exit Claude Code** - Type `exit` or press `Ctrl+D`
2. **Copy the command** shown above
3. **Paste and run** in your terminal (starts a recorded shell)
4. **Run `claude`** to start Claude Code inside the recording
5. Work normally - everything is captured
6. **Exit Claude Code** - Type `exit` or press `Ctrl+D`
7. **Exit the recording shell** - Type `exit` or press `Ctrl+D` again

Your recording will be saved to:
`$PWD/​tmp/​{workspace}_{datetime}.cast`

Phase 3: Additional Info

Purpose: Provide helpful tips for after recording.

## Tips

- **Environment variable**: `ASCIINEMA_REC=1` is set during recording
- **Playback**: Use `asciinema-player` skill or `asciinema play file.cast`
- **Upload (optional)**: `asciinema upload file.cast` (requires account)
- **Markers**: Add `asciinema marker` during recording for navigation points

TodoWrite Task Templates

Template: Record Claude Code Session

1. [Preflight] Check asciinema CLI installed
2. [Preflight] Offer installation if missing
3. [Context] Detect current workspace from $PWD
4. [Context] Generate datetime slug
5. [Context] Ensure tmp/ directory exists
6. [Command] Construct full recording command
7. [Guidance] Display step-by-step instructions
8. [Guidance] Show additional tips (playback, upload)
9. Verify against Skill Quality Checklist

Post-Change Checklist

After modifying this skill:

  1. Command generation still uses $PWD (no hardcoded paths)
  2. Guidance steps remain clear and platform-agnostic
  3. TodoWrite template matches actual workflow
  4. README.md entry remains accurate
  5. Validate with quick_validate.py

CLI Options Reference

OptionFlagDescription
Title-tRecording title (for asciinema.org)
Quiet-qSuppress status messages
Append-aAppend to existing recording

Troubleshooting

"Cannot record from within Claude Code"

Cause: asciinema must wrap the program, not be started from inside.

Fix: Exit Claude Code first, then run the generated command.

"Recording file too large"

Cause: Long sessions produce large files.

Fix:

  • Use asciinema upload to store online instead of locally
  • Split long sessions into smaller recordings

"Playback shows garbled output"

Cause: Terminal size mismatch.

Fix: Use -r flag during playback to resize terminal.


Reference Documentation

Post-Execution Reflection

After this skill completes, reflect before closing the task:

  1. Locate yourself. — Find this SKILL.md's canonical path before editing.
  2. What failed? — Fix the instruction that caused it.
  3. What worked better than expected? — Promote to recommended practice.
  4. What drifted? — Fix any script, reference, or dependency that no longer matches reality.
  5. Log it. — Evolution-log entry with trigger, fix, and evidence.

Do NOT defer. The next invocation inherits whatever you leave behind.

More skills from terrylica

code-hardcode-audit logo
terrylica/cc-skills

code-hardcode-audit

Detect hardcoded values, magic numbers, and leaked secrets. TRIGGERS - hardcode audit, magic numbers, PLR2004, secret scanning.

75 152
View
calcom-access logo
terrylica/cc-skills

calcom-access

Access Cal.com API via CLI with 1Password API key. Use when user wants to list bookings, create event types, manage schedules, or.

75 0
View
schema-e2e-validation logo
terrylica/cc-skills

schema-e2e-validation

Earthly E2E validation for YAML schema contracts. TRIGGERS - schema validation, YAML schema, schema contracts, regenerate types.

75 0
View
glossary-management logo
terrylica/cc-skills

glossary-management

Manage terminology glossary with Vale. TRIGGERS - sync terms, glossary validation, add terms, Vale vocabulary.

75 0
View
log-reader logo
terrylica/cc-skills

log-reader

Read MetaTrader 5 log files. TRIGGERS - MT5 logs, Experts pane, indicator errors, compilation errors.

75 0
View
daemon-start logo
terrylica/cc-skills

daemon-start

Start the asciinema chunker daemon. TRIGGERS - start daemon, resume chunker, enable backup.

75 0
View
sred-commit logo
terrylica/cc-skills

sred-commit

Create a git commit with SR&ED (CRA tax credit) trailers. TRIGGERS - sred commit, sred, cra commit, tax credit commit, scientific research commit.

75 0
View
doppler-workflows logo
terrylica/cc-skills

doppler-workflows

Manage credentials and secrets through Doppler for publishing and deployment workflows. Use whenever the user needs to publish Python packages.

75 0
View
distributed-job-safety logo
terrylica/cc-skills

distributed-job-safety

Concurrency safety patterns for distributed pueue + mise + systemd-run job pipelines.

75 0
View
diagnose logo
terrylica/cc-skills

diagnose

Diagnose Kokoro TTS issues. TRIGGERS - kokoro not working, tts diagnose, kokoro error, tts troubleshoot.

75 0
View
bootstrap-monorepo logo
terrylica/cc-skills

bootstrap-monorepo

Autonomous polyglot monorepo bootstrap meta-prompt on the moon + proto + Bun stack (Nx-convergent). TRIGGERS - new monorepo, new repository, polyglot setup, scaffold repo, moon proto bootstrap, monorepo from scratch.

75 0
View
fork-intelligence logo
terrylica/cc-skills

fork-intelligence

Discover valuable GitHub fork divergence beyond stars. TRIGGERS - fork analysis, fork intelligence, find forks

75 0
View

Popular AI tools

Kaiber logo
Video

Kaiber

Generate, edit, and beat-sync AI video with leading models in one workspace.

Paid
View
Vimcal logo
Productivity

Vimcal

The world's fastest calendar for remote work

Free
View

Transform Your Design with AI Designer by ImgCreator.ai

Freemium
View
Akool AI logo
Content & writing

Akool AI

Revolutionizing Video Production with AI-Powered Creativity

Paid
View

Extend an image past the frame and let AI fill the new aspect ratio.

Freemium
View
StarByFace logo
Security

StarByFace

Discover your celebrity doppelgänger with StarByFace!

Free
View
C

ChainClarity explains 700+ crypto whitepapers in plain English, with layered summaries, comparisons, research tools, alerts, and a $4.99 Pro plan.

Freemium
View
Opus Clip logo
Coding & apps

Opus Clip

Opus.ai: Revolutionize Your Web Experience

Free
View