pull-framework
Skill0 starsChanged 4 months agoMITView on GitHub
Project/Urgent/List/Limitations. Structured discovery before prescribing solutions. Understand the situation before acting. Activates on "what's the context", "tell me about the situation", "discovery", "understand before solving", "pull framework", or "what have you tried?"
In this repository15 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- ai-graduationSkill
- arbitrage-auditSkill
- assumption-mapperSkill
- b2b-idea-vettingSkill
- codifySkill
- decision-briefSkill
- design-thinking-liteSkill
- grill-meSkill
- grill-with-docsSkill
- innovation-diagnosticSkill
- kill-goSkill
- preflightSkill
- problem-framingSkill
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 4 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
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 14
Ask each in order. Don't prescribe solutions until all four are answered.
- Distinctive: line 22
If the user jumps to a narrow problem, zoom out first: "Before we dig into that — what's the bigger thing this is part of?"
- Distinctive: line 30
If nothing is urgent, that's useful information.
- Distinctive: line 50
Only after all four questions are answered, synthesize:
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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
The 4 Questions
Ask each in order. Don't prescribe solutions until all four are answered.
P — Project
"What are you working on?"
Get the full picture. Not just the task, but the context. What's the project? What's the goal? Who's involved? What stage is it at?
If the user jumps to a narrow problem, zoom out first: "Before we dig into that — what's the bigger thing this is part of?"
U — Urgent
"What's urgent about it?"
Why now? What's the deadline? What breaks if this doesn't get resolved? Is the urgency real (external deadline, user impact) or perceived (internal pressure, perfectionism)?
If nothing is urgent, that's useful information. It means there's time to think properly instead of reacting.
L — List
"What have you tried?"
Everything attempted so far. What worked partially? What failed completely? What did you learn from each attempt?
CRITICAL: People often skip what they've already tried and go straight to "help me solve this." Pull it out. Prior attempts contain signal about what the real problem is.
L — Limitations
"What's not working? What constraints are you operating under?"
What's blocking progress? Resource constraints? Technical limitations? Organizational politics? Knowledge gaps?
Separate constraints you can change from constraints you must accept. The strategy looks different for each.
After PULL
Only after all four questions are answered, synthesize:
- Restate the situation in your own words. Let the user correct you.
- Identify the actual problem (which may be different from what the user initially described).
- Propose a direction — not a full solution, but a next step.
Output
# Discovery Brief: [Topic]
Date: [date]
## Project
[What they're working on, in context]
## Urgency
[Why now, real vs. perceived]
## Prior Attempts
[What's been tried, what was learned]
## Limitations
[Changeable vs. fixed constraints]
## Actual Problem
[Restated after discovery — may differ from initial framing]
## Recommended Next Step
[One specific action]Based on the PULL Framework. Understand before prescribing.
Next step
Problem discovered → run problem-framing to diagnose deeper, or preflight if you're ready to scope a session.
Part of the ThinkHaven Method Kit by Kevin Holland. Full Board of Directors experience: ThinkHaven
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
Reviews can't be read right now.
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.