Review

Skill rulebased-io/claude-plugin/packages/second-brain/skills/review

Resurface and review old notes using spaced repetition or random selection. Use when the user wants to revisit their knowledge, process inbox items, or maintain note quality.From its SKILL.md

Install

npx -y skills add rulebased-io/claude-plugin --skill review

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

3.0 KB, 707 tokens by cl100k_base, as published. Nobody here has run it

Bring old notes back to attention for review, update, or archival.

When to Use

Regular review habit (weekly/daily)
Processing inbox backlog
Rediscovering forgotten knowledge
/rulebased-second-brain:review invoked

Procedure

Step 1: Select Review Mode

Ask user (or detect from arguments):

Inbox processing — Review unprocessed inbox items
Random resurface — Surface random notes from the brain
Oldest unreviewed — Notes with oldest or missing reviewed date
Tag-based — Review all notes with a specific tag

Step 2: Gather Review Candidates

Read AGENTS.md Structure table to know which folders to scan.

Inbox processing: list all files in inbox/, sorted by creation date (oldest first), batch of 5

Random resurface: glob **/*.md across content folders (exclude system/, templates/), randomly select 5, prefer notes not reviewed in 30+ days

Oldest unreviewed: scan frontmatter reviewed field, sort ascending (missing = highest priority), pick top 5

Step 3: Present Each Note

## Review: <title>
Created: <date> | Last reviewed: <date or "never">
Tags: <tags> | Status: <status>
Location: <file path>

<first 10 lines of content or full if short>

---
Actions: [keep] [update] [connect] [archive] [skip]

Step 4: Process User Action

Action	What happens
keep	Update `reviewed` timestamp, no other changes
update	User edits content, update `reviewed` and `updated`
connect	Invoke connect skill for this note
archive	Move to `archives/`
skip	Move to next note, no changes

For inbox processing, additional actions:

Action	What happens
promote	Move from `inbox/` to appropriate folder (read AGENTS.md to determine destination), set `status: seedling`
merge	Merge content into an existing note

Step 5: Maturity Check

After reviewing a note, evaluate whether its status should be promoted per system/conventions.md maturity model:

seedling with connections and refined content → suggest promoting to budding
budding with dense links and polished content → suggest promoting to evergreen

Step 6: Summary

Review complete:
- Reviewed: 5 notes
- Kept: 2 | Updated: 1 | Archived: 1 | Skipped: 1
- Maturity promotions: 1 (seedling → budding)
- Next review suggestion: 3 days

Rules

Always update the reviewed field with current date (YYYY-MM-DD).
When moving files (archive, promote), update all [[wiki-links]] that reference them.
Never delete notes during review. Archive instead.
Present notes one at a time, wait for user action before proceeding.
Keep the review flow fast — minimize unnecessary output.
Use folder paths from AGENTS.md, not hardcoded paths.

$ARGUMENTS

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Gives 0 of the 12 instructions most review quality skills give in 707 tokens

Counted across 1,048 of the 1,783 authors here whose files we hold, read 2026-08-07

Ask questions one at a timein 81 of 1048, across 64 files
Provide a recommended answer for each questionin 73 of 1048, across 50 files
Explore the codebase instead of asking answerable questionsin 66 of 1048, across 42 files
Resolve dependencies between decisions one-by-onein 42 of 1048, across 17 files
Interview the user relentlessly about the planin 38 of 1048, across 13 files
Order findings by severityin 31 of 1048
Resolve each branch of the decision treein 27 of 1048, across 5 files
Run a grilling sessionin 26 of 1048, across 5 files
Update CONTEXT.md immediately when a term is resolvedin 26 of 1048, across 11 files
Propose precise canonical terms for vague languagein 25 of 1048, across 7 files
Create documentation files lazilyin 24 of 1048, across 5 files
Assign severity to every findingin 24 of 1048

Said here and by no other author read

read structure table to find folders to scan
gather up to 5 review candidates based on mode
present notes one at a time
update reviewed field with current date
update wiki links when moving files
suggest status promotion when criteria are met

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.