oguzdogmus/mimir / skills/mimir-core-rules
mimir-core-rules
VerifiedHard rules for Mimir, the exam study plugin. Load this before any other Mimir skill runs and before reading, writing, or acting on a course workspace (the learning root, default ~/learning/, or any Mimir .state/ file).
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 20.8 KB |
| references/recording.md | markdown | 1.8 KB |
| references/session-close.md | markdown | 1.8 KB |
| references/verified-items.md | markdown | 4.0 KB |
references/recording.md
View on GitHub# Recording mastery
Call `python3 <mimir>/scripts/session.py record` with one JSON object on stdin:
`{"term":"2026-Fall","course":"CS201 Theory","topic":"Heaps","item":"paper.pdf:p3:q2","gate":{"decision":"show","counts_for_mastery":true,"label":"[past paper]"},"outcome":"unassisted","real_past_paper":true,"session":"unique-session-id","date":"2026-10-31"}`.
Use the actual gate result, stable item identity (same question keeps the same identity), and one unique session identity reused throughout this session. Use the date from `session.py now`. Outcomes are `unassisted | assisted | wrong`; Urgent drill correct answers are always assisted. Knowledge-gauge answers (core rules' Knowledge gauge) are the exception in both classes: right is always unassisted, wrong is wrong. Set real_past_paper only for an actual past-paper question, not generated practice or a worked example.
For a frustration override add `override: true` and use outcome assisted; this is a control event, not a counted solve. For an explanation without a gated item add `explanation_only: true` and pass decision drop and counts_for_mastery false. Retrying the same session/item/event is idempotent. Standard retries: record every try on an item. Within one session the first wrong answer on an item is recorded once (later misses return `duplicate`), and a right answer after it records as assisted, even when sent as unassisted.
Print or use the returned status; log refusals without inventing evidence. Subsequent answers in the override session do not count. A later unassisted answer lifts the block but still needs D2 evidence to promote.
Use `session.py check` or `progress` for current status and evidence. The Markdown file preserves legacy rows and notes; the recorder-owned state supersedes those rows. Only record changes mastery; initialization may create an empty table.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Unverified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |