- Home
- Skills
- Game Development
- First-Session Friction Audit
First-Session Friction Audit
Find where new players quit in the first ten minutes: measured, timestamped and screenshotted.
$5
Works with the AI Agents you already use
First-Session Friction Audit
Example session with this skill installed
Run a first-session friction audit on the latest build of 'Starshifter'. Use the skipper run method to see where players get stuck if they ignore the dialogue.
- Read your context and instructions
- Compiled the first-session friction audit
I have completed the friction audit for 'Starshifter' (Build v0.8.4) on Desktop. The audit identified three high-friction flags, specifically during the 'Skipper' run where the player bypassed the navigation tutorial.
First-Session Friction Audit: Starshifter
| Metric | Result | Flag Status | Evidence |
|---|---|---|---|
| Time to First Core Verb (Move) | 0:12 | Pass | Log: 00:12.04 |
| Time to First Core Verb (Shoot) | 4:45 | FLAG | Action gated by 3 unskipped popups |
| Forced Waits | 0:00 | Pass | No timers in first 10m |
| Unannounced Gates | 1 hit | FLAG | Refusal: "Need Plasma Core" at 03:20 |
| Reading Load (Pre-Action) | 142 words | FLAG | 5 popups before combat unlock |
| Travel Time | 18s | Pass | Hub to Hangar transit |
| Chores | 0 | Pass | No repeated actions > 30 |
| Crowding | 12 bots | Pass | Visibility clear in Hangar |
| Refusals/Errors | 0.2/min | Pass | 1 server refusal on zone transition |
Identified Friction Points
Unannounced Gate: At 03:20, the Skipper hit the Hangar door and received a "Need Plasma Core" refusal. The instruction for obtaining this core was contained in the skipped "Commander Vane" dialogue. No waypoint or HUD hint was provided post-skip.
Reading Load: The game requires dismissing 5 modal windows totaling 142 words before the 'Shoot' verb is enabled. This exceeds the benchmark of 60 words/3 popups.
Next steps
- Review the
evidence/folder for screenshots of the Hangar gate refusal. - Consider implementing a non-modal HUD objective marker for the Plasma Core to support "Skipper" players.
- Consolidate the initial 5 popups to get players to the combat tutorial under the 60-word threshold.
Connects securely to your tools. The creator never sees your data.
About this skill
Onboarding flows often leak players because the "time to fun" is buried under text walls, forced waits, and unannounced gates. This skill audits the first 10-60 minutes of your game to identify exactly where friction causes churn.
What it does
- Metric logging tracks timestamps for every core verb like moving, fighting, or crafting from a fresh account.
- Skipper simulation tests how the game handles players who dismiss every dialogue box to find where they get stuck.
- Reading load analysis counts word counts and popup frequency before the first core action is allowed.
- Crowd testing uses bots to measure hub congestion and visibility issues in starting areas.
- Evidence collection pairs every flagged friction point with timestamps, log lines, and screenshots.
How it works
- Setup begins with a fresh account and clean state on a specific build and device class.
- Execution runs a scripted "intended" path followed by a "skipper" run to find breakage points.
- Bot Injection populates hubs to test visibility and navigation under target concurrency.
- Report Generation produces a structured audit sheet using
templates/audit-sheet.mdwith flagged metrics.
Frameworks & tools
This skill works with existing bot harnesses, playtest frameworks, debug consoles, or manual screen recordings (30fps+) with timestamped event logs.
Why this beats prompting it yourself
Generic LLMs cannot simulate the specific "skipper" behavior or calculate reading load against game-design benchmarks. This skill enforces a rigorous measurement protocol that separates developer intent from actual player experience.
Use cases
- Auditing the First Time User Experience (FTUE) before an external playtest.
- Re-evaluating onboarding flow after a major tutorial or starting-area patch.
- Comparing your game's "time to first action" against reference titles in the genre.
- Identifying unannounced gates or chores that trigger early session abandonment.
Known limitations
This skill measures and reports friction but does not implement design changes or code fixes. It requires a fresh account for every run to ensure first-time-only flows are captured.
Does not implement design changes. Requires fresh accounts for every run. Needs a playtest harness, debug console, or screen recording to collect data.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 7 days ago