quick-sync
Skill2 starsChanged 7 months agoMITView on GitHub
Session-end synthesis, daily version. Processes inbox, logs decisions, enriches wiki pages, updates journal/NOW.md + TODO.md + DONE.md, appends log.md, commits. No deep rigor — fast and clean. Run at the end of every work session.
- Commits and pushes
The file
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 2 are distinctive, and 3 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.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 14
Call /ingest <path> on it
- Very specific: line 16
If user explicitly marks a note for orphaning (e.g., not relevant), set status: orphaned and skip
- Distinctive: line 18
If the inbox is empty, skip.
- Very specific: line 54
If a decision supersedes a prior one, add ## Superseded by: <new slug> to the old file.
- Distinctive: line 93
If there are no changes to commit, skip.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill 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
Synthesize the current session into the brain. Fast, clean, no deep review.
Steps
1. Process inbox
For every file in inbox/*.md with frontmatter status: raw (or no status):
- Call
/ingest <path>on it - On success, update its frontmatter to
status: ingested - If user explicitly marks a note for orphaning (e.g., not relevant), set
status: orphanedand skip
If the inbox is empty, skip.
2. Capture decisions from session
Scan the session conversation for statements of the form:
- "we decided ..."
- "going with ..."
- "let's [choose/go/pick] X"
- "X over Y because ..."
For each, write a decisions/<YYYY-MM-DD>-<slug>.md file with frontmatter:
---
title: <one-line title>
date: <ISO date>
status: decided
builds_on: [<prior decision slugs if any>]
touched: []
---
# Decision: <title>
## Context
<what prompted this>
## Decision
<what was decided>
## Consequences
<what changes>
## Builds on
- [[<prior decision>]] (if any)If a decision supersedes a prior one, add ## Superseded by: <new slug> to the old file.
3. Enrich wiki pages
Based on session topic, identify 1-3 wiki pages that should have new facts appended. For each, append under a new dated section:
## <YYYY-MM-DD>: <what was learned>
<paragraph, with inline [Source: ...] markers pointing to this session or a source file>
(enriched YYYY-MM-DD from session)Enrichment is additive only. Never edit existing paragraphs.
4. Update journal
journal/NOW.md— refresh this week's tracks, open decisions, blockers. UpdateLast updated:date.journal/TODO.md— move completed items tojournal/DONE.md. Add new tasks surfaced this session.journal/DONE.md— append items completed.
5. Append to log.md
## [YYYY-MM-DD] quick-sync
- Inbox processed: N
- Decisions logged: N
- Pages enriched: [list]
- Tasks closed: N6. Commit and push
git add -A
git commit -m "quick-sync: <short summary of session>"
git pushIf there are no changes to commit, skip.
Failure modes
/ingestfails on an inbox note → leave the note withstatus: raw, note in log.md, continue with rest- Git push fails → commit locally, print the error, let user resolve
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.

