Complete AI Training
Sign inGet my AI kit

Your job's AI kit

Get your AI kit

Tell us who you are and what you do. We show you your kit right away and email you the link: skills, prompts, AI agents, MCP servers and courses for your job.

500+ jobs ready, and we make a kit for any other job. No payment needed to look.

Share

AI agent for game developers

Quest and Narrative Dependency Check Agent

A quest set where every path from start to end resolves, with dead ends and unreachable content listed and fixed

Quality controlCreatives
See how it worksGet access
Quest and Narrative Dependency Check Agent: what goes in, what the agent does and what you get

What it does

A designer changes the item that unlocks a door in quest 3, and quest 9 quietly becomes impossible because it needed that door open. This agent reads the quest data, including steps, flags, required items, dialogue conditions and locations, and builds a graph of how each quest can be completed. It walks every path from start to end, listing dead ends, unreachable quests, missing items and flags that are never set. After a fix, it re-walks the graph and loops until all paths resolve. It also checks that each ending has at least one valid path and that no step can be skipped to break the story. The designer approves each change. Edge case: an optional side quest that can only start after the main story ends is flagged as an intended late quest if marked so in the data.

How it works

Follow the arrows from top to bottom. The orange dashed arrow is the loop: when a check fails, the agent goes back and tries again.

Start and resultWhat it doesA check on its own workWaits for your OKGoes back and retries
Yes, continueApprovedYes, continueNoNo 1 STARTS WHEN Quest data changes 2 USES A TOOL Read quest, flag, item and location data 3 DOES Build the dependency graph of steps, flags and items 4 DOES Walk every path from each start state to the endings 5 CHECKS THE RESULT Does every quest step have a path forward and a pathto reach it? If not: List dead ends and unreachable steps with themissing flag or item. Back to step 3. 6 DOES Propose a fix for each failure 7 YOU APPROVE Designer approves each change 8 USES A TOOL Re-read the updated data 9 CHECKS THE RESULT Do all paths resolve and does each ending still havea valid route? If not: Walk the changed quests again and list the newfailures. Back to step 3. 10 RESULT Dependency report with all paths verified
Read the steps as a list
  1. Quest data changes
  2. Read quest, flag, item and location data
  3. Build the dependency graph of steps, flags and items
  4. Walk every path from each start state to the endings
  5. Does every quest step have a path forward and a path to reach it?If not: List dead ends and unreachable steps with the missing flag or item. Back to step 3.
  6. Propose a fix for each failure
  7. Designer approves each changeThe agent waits here for your OK.
  8. Re-read the updated data
  9. Do all paths resolve and does each ending still have a valid route?If not: Walk the changed quests again and list the new failures. Back to step 3.
  10. Dependency report with all paths verified

How it decides

It walks the graph from every valid start state and flags any step that cannot be reached or has no way forward, excluding content marked as intentionally gated.

  • Flag a step as unreachable if no sequence of valid actions can set its required flags
  • Flag any flag that is read but never set
  • Treat content marked as gated as expected, not an error
  • Re-walk the whole graph after every change, not only the changed quest

Make it yours

Every agent is a starting point. You choose these settings for your own situation.

  • Data format and file location
  • Content marked as intentionally gated
  • Endings to verify
  • Flags to ignore
  • When the check runs

What keeps you in control

It always asks you first

  • Designer approves each change to the quest data

Hard limits

  • Never edit quest data itself
  • Never mark a path valid without walking it

It stops when

  • Done: All paths resolve and every ending is reachable
  • Stop: A cycle prevents the walk from finishing, so the agent reports the loop for the designer

Set it up

We guide you through the set-up, step by step

Members get the full set-up guide for this agent. No technical skills needed: you copy, paste and upload.

10 minto set it up in your AI
5 AIsChatGPT, Claude, Copilot, Gemini, Grok
  • One set of instructions to paste into your AI, with the clicks for ChatGPT, Claude, Microsoft 365 Copilot, Gemini and Grok
  • The agent then walks you through connecting your own data, one source at a time
  • A downloadable copy with the flow chart, the rules and the full guide
Get access to this agent

An example run

What happensAfter the designer moves the key from the cellar to the study, the walk shows quest 9, Smuggler's Debt, needs flag cellar_open that is no longer set. The agent proposes setting it when the study key is used. After the change, the re-walk finds quest 9 reachable but a dialogue branch with no exit. The agent lists it, the designer approves a fix, and the final walk is clean.

More agents for game developers

Creating · Creatives

Subdivision and Retopology Budget Agent

A low-poly asset inside the triangle budget with baked maps that match the source

9 steps2 checks2 jobs
Quality control · Creatives

Level Layout Playability Review Agent

A level with no unreachable areas or unplanned spikes, ready for playtest

10 steps2 checks2 jobs
Quality control · Creatives

Build Smoke Test Gatekeeper Agent

A build that has passed the smoke path or a clear report on what broke and why

10 steps2 checks4 jobs
Quality control · Creatives

Save Data Compatibility Agent

Every supported old save loads correctly in the new build

10 steps2 checks1 job