theory-of-constraints
Skill0 starsChanged 36 days agoMITView on GitHub
Use this skill when the user is struggling with a bottleneck, asking what to fix first, or deciding between competing improvements. Trigger on: \"reviews are taking forever\", \"CI/CD is slowing us down\", \"should we hire or automate?\", \"we can't ship fast enough\", \"the team is overwhelmed\", or any question about what's limiting throughput. Also trigger when someone is optimizing something that might not be the real constraint. This skill applies Goldratt's 5 Focusing Steps — use it rather than guessing at process fixes.
3 other projects use a version of this.
Adopted by
Most recently taken up by
- kasperjunge/skillstheir own version, first seen Nov 8, 2026
- marcos-sponton/frameworks-as-skillstheir own version, first seen Oct 13, 2026
- deciqAI/knowledge-skillstheir own version, first seen Jul 7, 2026
Other versions3 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 3
- deciqAI/knowledge-skills★ 13same name
- kasperjunge/skills★ 7same name
- marcos-sponton/frameworks-as-skills★ 1same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository13 other files it gives its agents.
- AGENTS.mdAGENTS.md
- 1on1Skill
- context-reviewSkill
- gmail-inboxSkill
- lamport-problem-statementSkill
- learning-labSkill
- logseq-db-querySkill
- nix-sandboxSkill
- obsidian-zettelkastenSkill
- peer-reviewSkill
- portkey-statusSkill
- proton-inboxSkill
- teams-extractorSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 24
Don't declare the constraint until you understand the system.
- Distinctive: line 32
If the user has metrics (cycle time, lead time, DORA), use them.
- Distinctive: line 54
Keep recommendations specific to the user's system.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Theory of Constraints Skill
Goldratt's TOC gives a principled, formula-backed way to decide what to fix — not by gut feel, but by measuring what's actually limiting the system's goal.
The three measures (always anchor analysis here)
- Throughput (T) — rate at which the system produces its goal (shipped features, revenue, deployments)
- Inventory (I) — work in progress: PRs open, tickets in flight, features built but unreleased
- Operating Expense (OE) — cost to keep the system running (headcount, infra, tooling)
Net Profit = T − OE. ROI = (T − OE) / I.
Improvement priority: ↑T first, then ↓I, then ↓OE. A local efficiency gain that doesn't move T is not an improvement — it's noise.
Step 0: Diagnose before prescribing (Jonah approach)
Don't declare the constraint until you understand the system. Ask questions that guide the user to see it themselves — this produces better analysis and ownership of the fix:
- "Where does work queue up or wait the longest?"
- "What's the one thing, if it went faster, would let everything else go faster?"
- "Where do you feel the pain most — is that also where work piles up, or somewhere else?"
- "What happens when demand spikes — what breaks first?"
- "Are there stages that finish fast but then just wait for the next stage?"
If the user has metrics (cycle time, lead time, DORA), use them. If not, the answers to the above are usually enough to identify the constraint. Only proceed to the 5 steps once you have a clear picture.
5 Focusing Steps
- Identify — name the primary constraint explicitly
- Exploit — squeeze maximum throughput from the constraint without new resources
- Subordinate — align everything else to serve the constraint; upstream stages should not overproduce
- Elevate — if exploitation isn't enough, invest to increase constraint capacity
- Repeat — once this constraint is broken, find the next one (don't coast)
See goldratt.md for detailed theory, examples from The Goal, and the Evaporating Cloud conflict resolution tool.
Output format
A TOC analysis should be concrete and actionable, not theoretical. Structure it as:
- The constraint — name it explicitly (e.g. "code review turnaround", "manual QA step")
- Exploit — what can be done right now to get more out of it without adding resources
- Subordinate — what should the rest of the system stop doing / slow down / change to support the constraint
- Elevate — only if exploitation is already maxed out: what investment would expand capacity
- Efficiency traps — call out any local optimizations that look good but hurt global throughput
Keep recommendations specific to the user's system. Avoid generic advice.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

