unreal-playtest-agent SKILL.md skillA
unreal-playtest-agent from dcc-mcp/dcc-mcp-unreal is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add dcc-mcp/dcc-mcp-unreal/unreal-playtest-agent@git:20260831.e5665e8Pins git:20260831.e5665e8 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- dcc-mcp/dcc-mcp-unreal
- Path
src/dcc_mcp_unreal/skills/unreal-playtest-agent/SKILL.md- Versions recorded
- 5
- Latest version
- git:20260831.9c079dd, committed Aug 31, 2026
- First version recorded
- Aug 25, 2026
- Size
- 8,388 bytes
- Repository stars
- 10
- License
- none declared
Domain skill - run screenshot-light PIE playtest episodes with structured entity observations, bounded semantic actions, transition polling, and in-memory traces for QA and external policy or RL runners.
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# Unreal Playtest Agent Use this skill to drive gameplay QA from Unreal Engine state instead of treating screenshots as the primary observation channel. ## Contract 1. Start an episode with bounded entity selectors and scalar attribute names. 2. Observe the active GameMode, controller, redacted possessed-pawn identity, UMG widgets, and selected runtime entities from the active PIE world. 3. Execute one finite semantic action. 4. Poll the action until it completes, stalls, or times out. 5. Evaluate project-specific rewards and oracles in a project-local Skill. 6. Finish the episode and retain the returned transition trace. ## Transition truth `runtime.possessed` is `null` unless the player controller directly reports a possessed pawn. When present, it contains the pawn name, class, whether it is the observation's player, an opaque hash of the object path, and a process- stable `sha256:` identity tied to the exact live actor object. The raw path is not exposed, missing paths never fall back to class and name, and a same-path replacement receives a different identity. Every episode binds one exact PIE world, session, player controller, and pawn. …
Read the whole file at its exact version.
How to install
mdr add dcc-mcp/dcc-mcp-unreal/unreal-playtest-agent@sha256:15d2488221bbdcdfThe line at the top of this page pins the label git:20260831.e5665e8, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
[](https://markdownregistry.com/a/art_ntpjr6f2slmt6zfw)
3 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 5 versions of this file, between Aug 25, 2026 and Aug 31, 2026, about one every 2 days.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20260831.9c079dd latest | 2026-08-31 | 9c079dd | 8,388 B | A | view · diff |
| git:20260831.e5665e8 | 2026-08-31 | e5665e8 | 6,891 B | A | view · diff |
| git:20260826.a028b91 | 2026-08-26 | a028b91 | 4,282 B | A | view · diff |
| git:20260826.4d4272c | 2026-08-26 | 4d4272c | 3,546 B | A | view · diff |
| git:20260825.6d2dc80 | 2026-08-25 | 6d2dc80 | 2,756 B | A | view |
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (8388 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Source
dcc-mcp/dcc-mcp-unreal · 10 stars · license none declared · pushed 2026-11-01 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_ntpjr6f2slmt6zfw GET https://markdownregistry.com/api/v1/resolve?ref=dcc-mcp/dcc-mcp-unreal/unreal-playtest-agent GET https://markdownregistry.com/api/v1/blob/15d2488221bbdcdfbf7e816bed6fc30b95336527461a1bbfe96d02f64c4d65db
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.