bottleneck-gating
Skill1 starsChanged 23 days agoMITView on GitHub
Phase plans by measured bottleneck, not by requested order. Each phase
In this repository46 other files it gives its agents.
- AGENTS.mdAGENTS.md
- .codex/AGENTS.mdAGENTS.md
- .gemini/AGENTS.mdAGENTS.md
- .cursor/rules/agent-foundry-orchestrator.mdcCursor rule
- .cursor/rules/agent-foundry-skills.mdcCursor rule
- web/llms.txtllms.txt
- anti-slopSkill
- api-designSkill
- automation-pickSkill
- budget-guardSkill
- cache-identitySkill
- constraint-then-solveSkill
- context-optimizationSkill
- cron-troubleshootSkill
- e2e-test-strategySkill
- engram-routingSkill
- execution-observabilitySkill
- fallback-model-routingSkill
- feedback-loopSkill
- judge-integritySkill
- knowledge-extractSkill
- landscape-firstSkill
- measure-firstSkill
- plan-before-codeSkill
- plan-then-actSkill
- prompt-disciplineSkill
- pushback-when-wrongSkill
- quality-protocolSkill
- read-before-buildSkill
- re-verify-findingsSkill
- session-closeoutSkill
- session-distillSkill
- show-your-workSkill
- structured-output-contractSkill
- verify-firstSkill
- workflow-decomposeSkill
- agent-memory-vaultSkill
- agent-team-standardsSkill
- autonomous-briefingSkill
- book-to-skillSkill
- chrome-devtools-mcp-bridgeSkill
- design-languageSkill
- funnel-pr-guardSkill
- persistent-memorySkill
- sql-migration-trioSkill
- token-compressionSkill
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 33
Ask: what is the single number that caps everything else?
- Distinctive: line 51
Every phase transition has a measurable condition:
- Distinctive: line 62
Every multi-phase plan must explicitly state:
≈ “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
Bottleneck Gating
The right sequence is: measure → fix the bottleneck → gate → fix the next bottleneck.
When to Use
Activates on any planning, roadmap, or execution task.
- Creating any multi-phase plan
- Asked to redesign before the core loop works
- Asked to bill before users get value
- Prioritizing a backlog
- Deciding what to build next
Procedure
1 — Find the bottleneck
Ask: what is the single number that caps everything else?
- 86% of cases die at step 1 → funnel is the bottleneck
- 3 users ever saw the core value → don't optimize that page yet
- 0 customers → don't build billing features
2 — Phase by what unlocks the next thing
Structure phases as:
Phase 0: Stop the bleeding (trust leaks, data, safety)
Phase 1: Fix the bottleneck (activate, convert, retain)
Phase 2: Monetize the working loop (only if Phase 1 succeeds)
Phase 3: Polish the experience (only if Phase 2 works)
Phase 4: Consolidate (only if later phases ship)3 — Define gates between phases
Every phase transition has a measurable condition:
Gate to Phase 2: case_created → reflection_submitted ≥ 50% for 2 consecutive weeks
If gate fails: investigate why, fix the cause, re-measure. Do not proceed.Gates protect against building on a broken foundation. If the gate isn't met, the next phase won't help.
4 — State the gate in the plan
Every multi-phase plan must explicitly state:
- The bottleneck being fixed in each phase
- The metric that proves the gate is met
- The trigger for moving to the next phase
Anti-patterns
Building billing features when no users complete the core flow. Building redesign when retention is the actual problem. Optimizing for power users when activation is the bottleneck.
- Ordering phases by stakeholder preference.
- Setting a gate with no metric or time window.
- Starting the next phase while a gate is failing.
Composes With
measure-first— provides the bottleneck measurementverify-first— confirms the gate metric actually moved
Verification Checklist
- One bottleneck caps the plan.
- Every phase has a measurable gate.
- Gate windows and thresholds are stated.
- A failed gate blocks the next phase.
- Phase 0 addresses trust, data, or safety first.
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.

