I Tested Superpowers v6 vs Grill Me: The Results Are Shocking
This video reruns one frozen feature through Superpowers v6 and a Grill-with-Docs workflow, then compares them with corrected earlier benchmarks for speed, tokens, cost, scope, and visible product defects. Superpowers won this measured run on efficiency and first-pass usability, while the repaired Grill result delivered broader functionality at much greater cost.
The Gray CatWatchTranscript found
Quick learning frame
Read this before watching.
A model becomes useful when it is wrapped in a harness: tools, state, permissions, memory, routing, and verification.
New playlist item from The Gray Cat; queued for transcript-backed review, topic mapping, and a practical learning artifact.
Skill you build: The ability to compare agentic coding workflows by controlling the task and weighing autonomy, implementation scope, token cost, review coverage, and working user paths together.
Watch for the shift from claim to mechanism. The learning value is the point where the transcript reveals a repeatable action, tool boundary, context move, review habit, or artifact.
Concept diagram
Where this video fits.
01User intent
02Model role
03Tool surface
04State and memory
05Verification loop
06Reusable operating rule
Deep lesson
Turn this video into working knowledge.
1,443 cleaned transcript words reviewed across 464 timed caption segments.
Thesis
I Tested Superpowers v6 vs Grill Me: The Results Are Shocking teaches a practical agent harness move: This video reruns one frozen feature through Superpowers v6 and a Grill-with-Docs workflow, then compares them with corrected earlier benchmarks for speed, tokens, cost, scope, and visible product defects. Superpowers won this measured run on efficiency and first-pass usability, while the repaired Grill result delivered broader functionality at much greater cost.
The goal is not to remember the video. The goal is to extract the operating principle, tie it to timestamped evidence, test how far the claim transfers, and make something reusable.
0:36
Control the Rerun
“corrected Poco and Osmani results. Quick reminder for anyone who missed my previous video. The gray cat code and real footage were unchanged. I reused the exact feature request and will leave it on the screen. Both main...”
Both new runs started from the same frozen product code and exact feature request, used Opus 5 at medium effort for the controller, and had Beats installed. Superpowers used its own planning and progress system, while the Poco plugin had changed by 32 commits, an important limitation when relating the rerun to earlier results. Write a benchmark preflight that records the frozen code revision, exact request, controller model, installed tools, and plugin revisions before any run starts.
2:26
Review Every Task
“focus on what changed, smaller tasks can also skip a separate design document. Their projects own tests report almost 50% fewer tokens at roughly twice the speed. Those are their results. I wanted to see how it handled...”
Superpowers v6 turned the approved design into nine small tasks, assigned each to a fresh implementer and one reviewer, and used a final Opus branch review that caught a repeated-sentence bug missed by the task reviews. Version 6 reduces duplicated review work by combining requirements and quality review and passing file paths instead of repeating large text blocks. Sketch the Superpowers loop from design approval through implementer, reviewer, fix cycle, final branch review, and verification.
6:08
Verify the Workflow
“less. Among 90% of Poc's cost went into the Orca implementation phase. Only three workers received their full implement skill. Poc also built more. It added 861 line of production code compared with 236 from Superpowers. Line count...”
In the Grill-with-Docs retry, only three of eight workers received the full Implement functions; the other five got the invocation as plain text, and the transcript explicitly confirms that ticket seven's POC review did not run. Separately, the player and timeline were missing until repair—the source does not establish that the skill-delivery problem caused those UI defects—while the repaired result delivered broader functionality at much greater time and cost than Superpowers. Create a launch checklist that confirms every worker loaded its required skill and that tests the complete visible user path before comparing feature breadth.
01
User intent
Start with this video's job: This video reruns one frozen feature through Superpowers v6 and a Grill-with-Docs workflow, then compares them with corrected earlier benchmarks for speed, tokens, cost, scope, and visible product defects. Superpowers won this measured run on efficiency and first-pass usability, while the repaired Grill result delivered broader functionality at much greater cost. Treat "User intent" as the outcome you are trying to make visible, not a topic label. Anchor it to 0:36, where the video says: “corrected Poco and Osmani results. Quick reminder for anyone who missed my previous video. The gray cat code and real footage were unchanged. I reused the exact feature request and will leave it on the screen. Both main...”
02
Model role
Use "Model role" to locate the part of the agent harness mechanism the video is demonstrating. Ask what changes in your real setup if this claim is true. Anchor it to 2:26, where the video says: “focus on what changed, smaller tasks can also skip a separate design document. Their projects own tests report almost 50% fewer tokens at roughly twice the speed. Those are their results. I wanted to see how it handled...”
03
Tool surface
Turn "Tool surface" into the reusable artifact for this lesson: A one-page agent harness map with tool boundaries, state ownership, and proof signals. This is where watching becomes something you can inspect and reuse.
04
State and memory
Use "State and memory" as the application surface. Decide whether the idea touches a browser flow, a local file, a model choice, a source document, a UI, or a review step.
05
Verification loop
Use "Verification loop" to prove the lesson. The evidence should connect back to the video title, transcript anchors, and a concrete output, not a generic best-practice claim.
06
Reusable operating rule
Use "Reusable operating rule" to carry the idea forward: save the prompt, checklist, diagram, or operating rule that would make the next agent run better.
Example
Source-backed artifact packet
Convert the video into a scoped artifact request that includes the transcript claim, mechanism, acceptance criteria, and proof. The output should be a one-page agent harness map with tool boundaries, state ownership, and proof signals..
Example
Agent harness proof brief
Separate what the speaker claims, what the demo actually proves, and what still needs outside verification before you adopt the agent harness pattern.
Example
Teach-back module
Transform the lesson into a definition, a User intent -> Model role -> Tool surface -> State and memory -> Verification loop -> Reusable operating rule diagram, one misconception, one practice exercise, and a check-for-understanding question.
Do not learn it wrong
Treating the title as the lesson without checking what the transcript actually says.
treating model choice as architecture
ignoring tool permissions
missing verification evidence
Letting the lesson drift into generic agent definitions.
Letting the lesson drift into model leaderboard claims.
Letting the lesson drift into tool list without operating boundaries.
Do not count this as learned until these are true.
01
State the transcript-backed claim in your own words: This video reruns one frozen feature through Superpowers v6 and a Grill-with-Docs workflow, then compares them with corrected earlier benchmarks for speed, tokens, cost, scope, and visible product defects. Superpowers won this measured run on efficiency and first-pass usability, while the repaired Grill result delivered broader functionality at much greater cost.
02
Explain the practical stakes without hype: New playlist item from The Gray Cat; queued for transcript-backed review, topic mapping, and a practical learning artifact.
03
Map the idea onto the User intent -> Model role -> Tool surface -> State and memory -> Verification loop -> Reusable operating rule sequence and name the weakest link.
04
Produce the artifact and include the evidence that proves it: A one-page agent harness map with tool boundaries, state ownership, and proof signals.
Put it into practice
Give this grounded prompt to Codex or Claude after watching.
You are helping me turn one specific YouTube video into real, durable learning.
Source video:
- Title: I Tested Superpowers v6 vs Grill Me: The Results Are Shocking
- URL: https://www.youtube.com/watch?v=d8lxLTQhmPo
- Topic: Agent Architecture
- My current learning frame: Run two agent workflows against one frozen feature, verify each worker's instructions before launch, and score time, all-agent tokens, implemented scope, review catches, and the complete user flow.
- Why this matters: New playlist item from The Gray Cat; queued for transcript-backed review, topic mapping, and a practical learning artifact.
Transcript anchors from this exact video:
- 0:36 / Evidence 1: "corrected Poco and Osmani results. Quick reminder for anyone who missed my previous video. The gray cat code and real footage were unchanged. I reused the exact feature request and will leave it on the screen. Both main..."
- 2:26 / Evidence 2: "focus on what changed, smaller tasks can also skip a separate design document. Their projects own tests report almost 50% fewer tokens at roughly twice the speed. Those are their results. I wanted to see how it handled..."
- 4:13 / Evidence 3: "disabled at first. After fixing that, the player and timeline disappeared. That comparison did not feel fair, so today I ran the full documented workflow on a new branch. I cleared the tracker first and started again. Grill..."
- 6:08 / Evidence 4: "less. Among 90% of Poc's cost went into the Orca implementation phase. Only three workers received their full implement skill. Poc also built more. It added 861 line of production code compared with 236 from Superpowers. Line count..."
- 8:20 / Evidence 5: "I did not pay those amounts directly. The product results were closer. Wayfinder needed a run button fix. Osmany review queue rendered empty and today's Pocock retry hit the editor until repair. Superpowers showed no visible defect in..."
Video-aware target:
- Prompt lane: Agent harness
- Mechanism to extract: Identify what surrounding harness makes the model more useful than chat alone.
- Artifact to produce: A one-page agent harness map with tool boundaries, state ownership, and proof signals.
- Artifact must include: model role; tools; state/memory; permission boundary; verification proof
Your task:
1. Use the transcript anchors above as the primary source packet. If you add outside context, label it clearly as outside context and keep it secondary.
2. Create a source-check table with columns: timestamp, claim, transcript support, what the demo proves, confidence, and what still needs verification.
3. Extract the actual teachable mechanism from the video: Identify what surrounding harness makes the model more useful than chat alone. Do not invent claims that are not supported by the title, lesson frame, or transcript anchors.
4. Build a reusable learning artifact: A one-page agent harness map with tool boundaries, state ownership, and proof signals.
5. Include:
- a plain-English definition of the core idea
- a diagram or structured model using this sequence: User intent -> Model role -> Tool surface -> State and memory -> Verification loop -> Reusable operating rule
- answers to these source questions: What does the video claim the agent can do? | What surrounding system makes that claim plausible? | What proof is shown instead of merely asserted?
- 3 concrete examples that apply the video idea to real agentic work, such as a repo-editing harness; a local research assistant; a recurring refresh agent
- 2 failure modes the video helps prevent, chosen from the transcript evidence and these likely risks: treating model choice as architecture; ignoring tool permissions; missing verification evidence
- a checklist for the next real workflow, focused on: tool boundaries, state ownership, done signal, recovery path
- one practical exercise with a clear done signal: Map one current coding workflow as a harness and mark the first missing proof signal.
6. Add a "learning transfer" section: what changes in my workflow tomorrow if I actually learned this?
7. Add a "source check" section that cites which transcript anchor supports each major takeaway.
Quality bar:
- Make this specific to "I Tested Superpowers v6 vs Grill Me: The Results Are Shocking", not a generic Agent Architecture essay.
- Tie each harness element to a transcript anchor that names a tool, state boundary, permission, model behavior, or verification step.
- Prefer operational examples, failure modes, and reusable artifacts over broad definitions.
- Call out uncertainty instead of smoothing over weak evidence.
- Avoid these generic drifts: generic agent definitions; model leaderboard claims; tool list without operating boundaries.
- If evidence is weak or missing, stop and say what transcript segment or timestamp needs review instead of guessing.
- Finish with a concise artifact I could paste into my learning app.
Misconceptions
What to stop believing.
A better model automatically makes a better agent.
The model matters, but harness design determines whether the system can act safely and repeatably.
More tools always help.
Every tool increases surface area. Strong agents have the right tools with clear permissions.
Memory means saving everything.
Useful memory is compressed, curated, and tied to future decisions.
Practice studio
Learning only counts when you make something.
01
Transcript evidence map
Separate what the video actually says from what you already believe about the topic.
3 source-backed takeaways with timestamps, confidence, and a transfer note.02
One useful artifact
Apply the video to a real workflow and produce a one-page agent harness map with tool boundaries, state ownership, and proof signals..
A reusable artifact with a done signal and one verification step.03
Agent harness teach-back card
Explain the agent harness mechanism to someone who has not watched the video yet.
A 90-second explanation, one diagram, one example, and one misconception to avoid.
Recall check
Answer first, then reveal — without rewatching.
Which major variables were held constant across the two new runs?
What did the final Superpowers branch review catch that nine task reviews missed?
What did result collection verify about Implement delivery and review coverage?
Source shelf
Use the video as a doorway, then verify with primary sources.
