agentsclimarketplace

Monthly compilation

Skill conside-ray/llm-journaling-system/skills/monthly-compilation

An Obsidian-based, agent-assisted journaling system that compiles daily notes into actionable reviews and durable personal knowledge.

Install
npx -y skills add conside-ray/llm-journaling-system --skill monthly-compilation

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

  • 1 stars1 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.

What its author says it does

Copied from the file, not written here

Compile weekly reviews into a strategic monthly review and consolidate promotion candidates using cross-week evidence.

SKILL.md

7.7 KB, as published. Nobody here has run it

Monthly Compilation

Purpose

Use this skill to compile one month of weekly reviews into:

  1. a draft monthly review in vault/journals/monthly/
  2. a refined update to vault/promotion/candidates.md

This is a higher-level synthesis workflow than weekly compilation.

Its goal is to identify:

  • what kind of month this really was
  • what repeated across weeks strongly enough to matter
  • where behavior aligned or drifted from stated priorities
  • which promotion candidates should be confirmed, refined, merged, deferred, or discarded

Inputs

Read the following inputs before drafting outputs:

Required

  • all weekly reviews for the target month, which are located under vault/journals/weekly/YYYY. For example, weekly reviews in 2026 are stored in vault/journals/weekly/2026.
  • vault/context/priorities.md

Use when relevant

  • relevant project files under vault/context/projects
  • relevant area files under vault/context/areas
  • relevant note file under vault/context/notes/
  • existing vault/promotion/candidates.md
  • canonical promotion files for deduplication awareness only
  • previous monthly review, if useful for continuity

Do not expand the scope unnecessarily. Prefer the smallest relevant context set that improves judgment.

Output Files

This skill produces two outputs:

  1. a monthly review draft in vault/journals/monthly/YYYY. For example, the monthly review for April 2026 should be stored in vault/journals/monthly/2026 and follow the naming convention of Monthly Review @yyyy-mm.
  2. an updated vault/promotion/candidates.md

This skill may update vault/promotion/candidates.md, but it must not directly update:

  • vault/promotion/insights.md
  • vault/promotion/ideas.md
  • vault/promotion/principles.md
  • files under vault//context

Core Workflow

Step 1: Read the month

Read the target month’s weekly reviews and identify:

  • recurring wins and supportive conditions
  • recurring frictions, drifts, and bottlenecks
  • repeated project and area signals
  • evidence of alignment or misalignment with stated priorities
  • patterns that appear across multiple weeks, not just once

Step 2: Read current priorities

Read vault/context/priorities.md and identify:

  • the current season
  • relevant monthly priorities
  • relevant quarterly priorities
  • declared drift or failure modes
  • explicitly deferred or non-current focuses

Use this file as the benchmark for alignment.

Step 3: Pull in relevant context

Read only the project and area files that materially improve interpretation.

Use project notes to understand:

  • intended outcomes
  • current bottlenecks
  • risks and fake progress modes
  • strategic significance

Use area notes to understand:

  • standards
  • current focus
  • recurring risks
  • positive and negative signals

Read notes under vault/context/notes/ only when they are explicitly linked in relevant source material or clearly necessary to interpret a repeated pattern, project signal, or area signal at monthly scale.

Use them selectively as supporting evidence or clarification. Do not expand monthly scope by scanning notes broadly.

Step 4: Review promotion staging

Read the current promotion/candidates.md and assess:

  • which candidates now look stronger
  • which candidates overlap and should be merged
  • which candidates remain weak or thinly evidenced
  • whether any higher-order candidate emerges only at monthly scale

Do not treat all prior candidates as valid by default.

Step 5: Diagnose the month

Form a grounded view of the month.

Answer:

  • What kind of month was this, really?
  • What moved meaningfully?
  • What remained stagnant?
  • What repeated strongly enough to matter?
  • Where did behavior align with stated priorities?
  • Where did it drift?
  • What area or project signals stand out at monthly scale?
  • Which promotion candidates deserve more confidence, less confidence, merging, or removal?

Step 6: Draft the monthly review

Generate the monthly review using the monthly review template located in templates/monthly-review.md.

Write clearly and diagnostically. Focus on:

  • what defined the month
  • what repeated
  • what this month revealed more clearly than a single week could
  • what should change next month

Step 7: Refine promotion candidates

Update vault/promotion/candidates.md based on monthly evidence.

Use the monthly layer to:

  • confirm strong candidates
  • refine wording
  • merge overlapping candidates
  • defer weak ones
  • discard stale or unsupported ones
  • add new candidates only when they clearly emerge at monthly scale

Promotion Rules

Monthly promotion work is stricter than weekly promotion work.

The monthly layer should prioritize:

  • confirmation
  • refinement
  • consolidation
  • pruning

over raw candidate generation.

Candidate quality bar

A candidate is stronger at monthly scale when it is:

  • visible across multiple weeks
  • strategically meaningful
  • behaviorally actionable
  • transferable beyond the immediate month

Monthly-specific actions

Use the monthly run to:

  • raise confidence on well-supported candidates
  • lower confidence on weak candidates
  • merge duplicate or overlapping items
  • sharpen vague candidates into cleaner statements
  • discard candidates that no longer look useful

New monthly candidates

Only add a new candidate if it becomes clear mainly from cross-week synthesis.

Examples:

  • a month-long drift pattern
  • a repeated area-level sacrifice
  • a broader principle that no single week clearly showed
  • a strategy or content idea that crystallized across several reviews

Do not promote

Do not strengthen or add candidates for:

  • isolated weekly frustrations
  • thin patterns with weak evidence
  • generic motivational ideas
  • inflated conclusions from a noisy month

Writing Standards

For the monthly review

Write like a strategic diagnostic memo.

The review should be:

  • concise
  • comparative
  • pattern-aware
  • action-oriented

Avoid:

  • rehashing daily detail
  • merely stacking weekly summaries
  • vague self-improvement language
  • false precision

For promotion candidate updates

Be conservative and explicit.

If refining a candidate:

  • make it clearer
  • make it more evidence-based
  • make it easier to route and review

If merging candidates:

  • preserve the strongest version
  • remove redundant phrasing

Consistency Rules

The monthly review and the candidate updates must agree.

Do not confirm a candidate at high confidence if the monthly review does not support it.

Do not let the monthly review ignore a strong repeated signal that is being promoted.

The monthly promotion stream should feel like a natural consequence of the monthly diagnosis.

Update Boundaries

This skill may:

  • create or update the monthly review draft
  • create or update vault/promotion/candidates.md

This skill must not:

  • edit daily journals
  • edit weekly reviews except by explicit instruction
  • edit files under vault//context
  • edit canonical promotion files
  • silently redefine priorities, projects, or areas

If a context file appears stale or misaligned, note it in the monthly review rather than rewriting it.

Completion Checklist

Before finishing, confirm:

  • the monthly review reflects the target month accurately
  • priorities were actually used as an alignment benchmark
  • repeated patterns were identified across weeks, not just copied from one week
  • candidate updates are more selective and refined than weekly outputs
  • weak candidates were not preserved automatically
  • no canonical promotion files were modified
  • no vault/context files were modified
  • the review and candidate updates are consistent with each other

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.