Master-Planner / rules
Cursor rule0 starsChanged 5 months agoMITView on GitHub
Keeps durable planning memory on disk and enables recovery after context loss or compaction
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/003-anti-rationalization.mdcCursor rule
- doubt-researchSkill
- intake-interviewSkill
- master-plannerSkill
- plan-grillerSkill
- project-memorySkill
- security-rollbackSkill
- spec-drafterSkill
- task-decomposerSkill
The file
We split this 004-memory-and-resume.mdc into the rules it gives an agent
6 rules. About 0 are common, 2 are distinctive, and 4 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 12
When a .cursor/plans/<project>/ directory exists:
- Very specific: line 13
Read manifest.md before continuing any in-progress plan.
- Very specific: line 14
Read context.md (the memory bank) before proposing architecture (check Key Decisions) and before re-fixing a bug (check Known Pitfalls).
- Very specific: line 17
Update manifest.md (phase, status, active assumptions, resume instruction) whenever the plan advances a phase.
- Distinctive: line 19
Never store secrets (keys, tokens, passwords, PII) in any memory or planning file.
- Distinctive: line 22
If you sense you have lost the thread (after compaction, a new session, or a long conversation):
≈ “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: Keeps durable planning memory on disk and enables recovery after context loss or compaction globs: * alwaysApply: true --- # Memory and Resume ## Why this rule exists The agent's chat context is volatile: it is lost on compaction, new sessions, and long conversations. Anything that must survive belongs on disk. This rule makes the file system the source of truth for planning state. ## Read memory before deciding When a `.cursor/plans/<project>/` directory exists: - Read `manifest.md` before continuing any in-progress plan. - Read `context.md` (the memory bank) before proposing architecture (check **Key Decisions**) and before re-fixing a bug (check **Known Pitfalls**). Do not re-litigate a settled decision recorded as an ADR. ## Write memory at phase boundaries - Update `manifest.md` (phase, status, active assumptions, resume instruction) whenever the plan advances a phase. - Append to `context.md` when a decision, fact, or pitfall emerges that will prevent a future mistake. Keep entries terse. Do not create memory churn — memory is a safety net, not a diary. - Never store secrets (keys, tokens, passwords, PII) in any memory or planning file. ## Recover after context loss If you sense you have lost the thread (after compaction, a new session, or a long conversation): 1. Read `.cursor/plans/<project>/manifest.md` first. 2. Then read `spec.md` and `tasks.md`. 3. Continue from the phase recorded in the manifest. Do not restart from scratch and do not invent state. ## Anti-Rationalization | Excuse | Reality | |--------|---------| | "I'll remember it" | Compaction will erase it. Write it down. | | "Updating the manifest is busywork" | It is the only thing that lets a future session resume. 20 seconds now saves a restart. | | "This decision is obvious, no need to record" | Obvious now, invisible after compaction. One ADR line prevents a re-litigation. |
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.