agentleFS
Liam-codingLiam-coding/SelfOS/.claude/skills/digest/SKILL.md

digest

Skill4 starsChanged 4 days agoMITView on GitHub

Use this skill when the user says "/digest", "process my sessions", "read the digest", "distribute insights", or "update vault from sessions". Reads the latest session digest from vault/90_Inbox/ and distributes insights across vault modules via the staging workflow.

36 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use aeon-update, auto-workflow and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • aaronjmarsSkill

    aeon-update

    Pull framework updates from the upstream Aeon repo into this instance - 3-way merges canon's new commits into a PR, never clobbering operator config.

    Also in 4 projects that use this

  • aaronjmarsSkill

    auto-workflow

    Two-mode aeon.yml workflow builder - analyze inspects URLs and emits a tiered, signal-verified skill-enablement plan plus an aeon.yml diff; enable flips slugs to enabled:true and opens a PR.

    Also in 4 projects that use this

  • +10 more from aaronjmars

The file

We split this skill into the rules it gives an agent

7 rules. 0 are common, 5 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 32 edited versions we read keep these.

  • 5

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Read the most recent *-sessions-digest.md file.

    line 29
  2. Distinctive:

    If no digest exists, tell the user:

    line 31
  3. Distinctive:

    Before writing anything, read:

    line 39
  4. Distinctive:

    Only write what is clearly supported by the digest text

    line 65
  5. Distinctive:

    Do not fabricate or extrapolate beyond what was said

    line 66
  6. Distinctive:

    Write one file per module that has new information.

    line 73
  7. Very specific:

    Commit with message: digest: distribute session insights YYYY-MM-DD

    line 113

“Common” is measured: kept by at least 6 in 10 of the 32 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

digest — Distribute Session Insights to Vault

What this does

Reads your latest session digest (written by harvest.py) and distributes what it learns about you across the vault modules where it belongs: career, work, research, goals, decisions, and current priorities.

Nothing is written without your approval.


Step 1 — Find the latest digest

ls vault/90_Inbox/ | sort | tail -5

Read the most recent *-sessions-digest.md file.

If no digest exists, tell the user:

"No digest found. Run python -m selfos harvest first, then come back." Stop here.


Step 2 — Read existing vault state

Before writing anything, read:

  • vault/NOW.md
  • vault/GOALS.md
  • vault/PROFILE.md (if exists)

This prevents overwriting what's already confirmed.


Step 3 — Analyze the digest

Work through every context section in the digest. For each one, extract:

What you findWhere it goes
Job title, employer, tech skills, GitHub repos01_Career/profile.md
Active projects, stack, tasks, blockers, next actions02_Work/[project].md
Research papers, experiments, hypotheses, results, venues03_Research/[project].md
Training, health, fitness04_Health_Fitness/log.md
Learning goals, courses, books05_Life/learning.md
Current priorities, what's active nowNOW.md
Explicit goals the user statedGOALS.md
Explicit decisions the user madeDECISIONS.md (append-only)

Rules:

  • Label every inference [ai-inference]
  • Only write what is clearly supported by the digest text
  • Do not fabricate or extrapolate beyond what was said
  • Include a short quote from the digest as evidence for each key claim

Step 4 — Write staging files

Write one file per module that has new information.

File naming: vault/91_Staging/YYYY-MM-DD-digest-[module].md

Each file starts with:

---
type: [career | work | research | goals | now | decisions | life | health]
updated: YYYY-MM-DD
source: session-digest
target: vault/[path to canonical file]
---

Step 5 — Present summary and wait

Show the user a table before touching anything:

Module             | File                                   | Key findings
-------------------|----------------------------------------|---------------------------
01_Career          | 91_Staging/...-digest-career.md        | [2-3 word summary]
02_Work/[project]  | 91_Staging/...-digest-work-[proj].md   | [2-3 word summary]
03_Research        | 91_Staging/...-digest-research-....md  | [2-3 word summary]
NOW.md             | 91_Staging/...-digest-now.md           | [2-3 word summary]

Then say:

"Review the staging files above. Say apply to write them to the vault, or tell me what to change."

Do not apply anything until the user explicitly approves.


Step 6 — Apply and commit

After approval:

  1. Apply each staging file to its target path
  2. Commit with message: digest: distribute session insights YYYY-MM-DD

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.