Master-Planner / rules
Cursor rule0 starsChanged 5 months agoMITView on GitHub
Prevents AI self-justification of weak planning
Where files are headingNew Cursor rule files have more rules about planning and tracking tasks, and fewer about code style.
Of the Cursor rule files we read, the 4,183 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.
In this repository12 other files it gives its agents.
- .cursor/rules/000-harness-bootstrap.mdcCursor rule
- .cursor/rules/001-planning-safety.mdcCursor rule
- .cursor/rules/002-source-discipline.mdcCursor rule
- .cursor/rules/004-memory-and-resume.mdcCursor rule
- doubt-researchSkill
- intake-interviewSkill
- master-plannerSkill
- plan-grillerSkill
- project-memorySkill
- security-rollbackSkill
- spec-drafterSkill
- task-decomposerSkill
The file
We split this 003-anti-rationalization.mdc into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
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
When you catch yourself about to skip a step, simplify a requirement, or declare something "good enough" — stop and check this table.
≈ “Common” is an estimate: at least 1 in 5 of the 6,668 Cursor rule 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
--- description: Prevents AI self-justification of weak planning globs: * alwaysApply: true --- # Anti-Rationalization ## Why This Rule Exists The #1 failure mode of AI-assisted planning is self-justification. The agent produces a weak plan, then rationalizes why it's fine. This rule exists to break that loop. Source: addyosmani/agent-skills (60k+ stars, MIT) — every skill includes an Excuse | Reality table. Adapted for planning context. ## The Rule When you catch yourself about to skip a step, simplify a requirement, or declare something "good enough" — stop and check this table. ## Planning Anti-Rationalization | Excuse | Reality | |--------|---------| | "Tests aren't needed for this" | If it ships, it needs tests. Write the test plan. | | "Security isn't relevant here" | If it handles user input, auth, or data, security is relevant. | | "The architecture is straightforward" | Then documenting it is fast. Write the ADR. | | "This is a v1, we can add that later" | Decisions deferred are decisions made poorly. Document what's deferred and why. | | "The user didn't ask for this level of detail" | The user asked for a reliable plan. That includes the necessary detail. | | "I already reviewed this in my head" | Generator ≠ Evaluator. If you wrote it, you can't review it. Call the griller. | | "The requirements are clear enough" | Clear to whom? If there are no acceptance criteria, they're not clear. | | "This dependency is safe, everyone uses it" | Popularity ≠ safety. Check the license, maintenance, and permissions. | | "We can refactor later" | Refactoring is the tax on skipped planning. Pay now or pay more later. | | "The plan is done, let's just start coding" | Is it approved? Is it grilled? Are there unresolved `[NEEDS CLARIFICATION]` markers? | ## Red Flags — Stop and Recheck - Using "should", "probably", "seems like", "I think" in planning artifacts - Skipping acceptance criteria on tasks - Marking a plan as complete without user approval - Claiming "production-ready" without a test plan, security review, or deployment strategy - Reviewing your own plan and finding zero issues
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.