agentleFS
Ar9avAr9av/obsidian-wiki/skills/daily-update/SKILL.md

daily-update

Skill3.7k starsChanged 9 days agoMITView on GitHub

Run or configure the daily wiki maintenance cycle: check source freshness, refresh the index and hot.md, and manage its scheduled 9 AM launchd/systemd/cron reminder. Use for daily or morning wiki refresh or scheduler setup and verification.

  • Reads credentials

7 other projects use a version of this, 1 of them edited.

Adopted by

Steady: 0 new projects took it in the last 4 weeks, none the 4 weeks before.thin data

Week of Sep 13: 0 new projectsWeek of Sep 20: 0 new projectsWeek of Sep 27: 0 new projectsWeek of Oct 4: 0 new projectsWeek of Oct 11: 0 new projectsWeek of Oct 18: 0 new projectsWeek of Oct 25: 0 new projectsWeek of Nov 1: 0 new projects
Sep 2026Oct 2026Nov 2026
New projects taking a version of this file each week, over the 8 weeks our latest run read: only 0 in all, so this is a hint, not a trend.

Most recently taken up by

Who's talking about Ar9av/obsidian-wiki

The biggest 5 of 27 mentions linking to the repository that our crawl found. Each line is their own headline.

See all 27

Talked about on GitHub and Bluesky (27 times).

On GitHub · 25

20 more on GitHub

On Bluesky · 2

The file

We split this skill into the rules it gives an agent

12 rules. 0 are common, 4 are distinctive, and 8 are very specific to this project.

  • 0

    Common

    Most of the 7 edited versions we read keep these.

  • 4

    Distinctive

    This author's own choices. Worth a look.

  • 8

    Very specific

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

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

    If either command reports the vault is unmigrated, stop and tell the user to run obsidian-wiki memory migrate (preview) then --apply.

    line 51
  2. Very specific:

    Write to the vault-scoped $STATE_DIR derived in "Before You Start":

    line 55
  3. Very specific:

    If the schedule says to run, invoke the wiki-lint skill, then record the run:

    line 77
  4. Very specific:

    After the cycle, spawn impl-validator as a subagent:

    line 87
  5. Very specific:

    Check that $OBSIDIAN_WIKI_REPO/scripts/daily-update.sh exists and is executable.

    line 134
  6. Very specific:

    Write the literal repo path in place of $OBSIDIAN_WIKI_REPO — cron does not load your shell env.

    line 171
  7. Distinctive:

    Ask the user: "Do you want a terminal reminder when your wiki is stale?

    line 175
  8. Distinctive:

    If yes, detect the user's shell and target the right rc file:

    line 177
  9. Very specific:

    Check if wiki-notify.sh is already sourced in that rc file.

    line 188
  10. Very specific:

    Use $QMD_CLI if set; otherwise use qmd.

    line 225
  11. Distinctive:

    If the output says vectors are needed or embeddings may be stale, run:

    line 231
  12. Distinctive:

    Verify the collection with either:

    line 237

“Common” is measured: kept by at least 6 in 10 of the 7 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

Daily Update — Wiki Maintenance Cycle

You run a lightweight maintenance pass over the wiki: check source freshness, refresh the index, update hot.md, and write the state file that the terminal notification reads.

Before You Start

  1. Resolve config — follow the Config Resolution Protocol in llm-wiki/SKILL.md (inline @name override → walk up CWD for .env → global config → prompt setup). This gives OBSIDIAN_VAULT_PATH and OBSIDIAN_WIKI_REPO.
  2. Derive vault-scoped state dir — all runtime state is scoped to the resolved vault, not global:
    VAULT_ID=$(echo "$OBSIDIAN_VAULT_PATH" | md5sum 2>/dev/null | cut -c1-8 || md5 -q - <<< "$OBSIDIAN_VAULT_PATH" | cut -c1-8)
    STATE_DIR="$(obsidian_wiki_config_dir)/state/$VAULT_ID"
    mkdir -p "$STATE_DIR"
  3. Read $OBSIDIAN_VAULT_PATH/.manifest.json.

Modes

Run Mode (default — triggered by cron or /daily-update)

Execute the maintenance cycle:

Step 1: Source freshness check

Compare each source in .manifest.json against its file's modification time. Classify as:

  • Fresh — mtime ≤ ingested_at
  • Stale — mtime > ingested_at (new content exists, not yet ingested)
  • Missing — source file no longer exists

Step 2: Index refresh

obsidian-wiki memory index --vault "$OBSIDIAN_VAULT_PATH"

This reconciles index.md against the pages on disk under the memory lock — missing entries added, entries for deleted pages removed, the owner's own sections left untouched. Note added/removed from the output for the log line in Step 6. Do not enumerate pages with find and edit the index by hand.

Step 3: hot.md update

obsidian-wiki memory hot --vault "$OBSIDIAN_VAULT_PATH"

Recent Activity, Active Threads, and Flagged Contradictions are regenerated from the log, the todo index, and page frontmatter; ## Key Takeaways carries across unchanged. If the takeaways are older than ~48h and the vault has changed materially since, refresh them: read the 10 most recently updated pages and pass a fresh ~500-word snapshot with --takeaways - on stdin. Otherwise leave them — a rebuild without new takeaways is cheap and correct.

If either command reports the vault is unmigrated, stop and tell the user to run obsidian-wiki memory migrate (preview) then --apply; do not fall back to hand-editing.

Step 4: Write state

Write to the vault-scoped $STATE_DIR derived in "Before You Start":

date +%s > "$STATE_DIR/.last_update"
echo "<stale_count>" > "$STATE_DIR/.pending_delta"
echo "$OBSIDIAN_VAULT_PATH" > "$STATE_DIR/.vault_path"

Step 4a: Scheduled health check (wiki-lint)

LINT_SCHEDULE (default weekly) controls how often this cycle also runs wiki-lint:

  • manual — never auto-run; skip this step entirely.
  • daily — run wiki-lint every cycle.
  • weekly — run wiki-lint only if $STATE_DIR/.last_lint is missing or older than 7 days.
LINT_SCHEDULE="${LINT_SCHEDULE:-weekly}"
NOW=$(date +%s)
LAST_LINT=$(cat "$STATE_DIR/.last_lint" 2>/dev/null || echo 0)

If the schedule says to run, invoke the wiki-lint skill, then record the run:

date +%s > "$STATE_DIR/.last_lint"

Fold its summary (broken links, orphans, stale pages found) into Step 7's report as a Health check: line; omit the line entirely on a cycle where lint didn't run.

Step 5: Spawn impl-validator

After the cycle, spawn impl-validator as a subagent:

impl-validator check:
  goal: "Daily wiki maintenance — index reconciled, hot.md refreshed, state file written"
  artifacts:
    - $OBSIDIAN_VAULT_PATH/index.md
    - $OBSIDIAN_VAULT_PATH/hot.md
    - $STATE_DIR/.last_update
    - $STATE_DIR/.pending_delta
  checks:
    - Does .last_update contain a recent Unix timestamp (within the last 60 seconds)?
    - Does .pending_delta contain a non-negative integer?
    - Does hot.md have an updated: frontmatter field set to today?
    - Does index.md list at least as many pages as exist in the vault?

Apply any FAILs before logging.

Step 6: Log

Append to $OBSIDIAN_VAULT_PATH/log.md:

obsidian-wiki memory log DAILY-UPDATE fresh=<N> stale=<N> missing=<N> index_added=<N> hot_refreshed=<true|false> lint=<ran|skipped>

Step 7: Report to user

## Daily Wiki Update

- Sources: N fresh · N stale · N missing
- Index: N pages (N added, N removed)
- hot.md: refreshed / up to date
- Health check: N broken links, N orphans, N stale pages (omit this line if lint didn't run this cycle)

Stale sources (run to sync):
  /wiki-history-ingest claude   — N sessions since last ingest
  /wiki-history-ingest codex    — N sessions since last ingest

Setup Mode (triggered by "set up the daily cron" or "install terminal notification")

Walk the user through first-time setup:

Step 1: Verify script exists

Check that $OBSIDIAN_WIKI_REPO/scripts/daily-update.sh exists and is executable. If not, point the user to it.

Step 2: Install the scheduler — pick by platform (uname -s).

macOS (Darwin) — launchd:

# Replace placeholder in plist
sed "s|OBSIDIAN_WIKI_REPO|$OBSIDIAN_WIKI_REPO|g" \
  "$OBSIDIAN_WIKI_REPO/scripts/com.obsidian-wiki.daily-update.plist" \
  > "$HOME/Library/LaunchAgents/com.obsidian-wiki.daily-update.plist"

# Load it
launchctl load "$HOME/Library/LaunchAgents/com.obsidian-wiki.daily-update.plist"

Linux with systemd (systemctl --user works) — a user timer:

UNIT_DIR="${XDG_CONFIG_HOME:-$HOME/.config}/systemd/user"
mkdir -p "$UNIT_DIR"
sed "s|OBSIDIAN_WIKI_REPO|$OBSIDIAN_WIKI_REPO|g" \
  "$OBSIDIAN_WIKI_REPO/scripts/obsidian-wiki-daily-update.service" \
  > "$UNIT_DIR/obsidian-wiki-daily-update.service"
cp "$OBSIDIAN_WIKI_REPO/scripts/obsidian-wiki-daily-update.timer" "$UNIT_DIR/"
systemctl --user daemon-reload
systemctl --user enable --now obsidian-wiki-daily-update.timer

On a headless server, user timers only run while the user is logged in unless lingering is on — suggest sudo loginctl enable-linger "$USER".

Anything else (no systemd, containers, WSL without systemd) — crontab. Append this line via crontab -e, skipping it if an obsidian-wiki daily-update line is already there:

0 9 * * * /bin/bash "$OBSIDIAN_WIKI_REPO/scripts/daily-update.sh" >> /tmp/obsidian-wiki-daily.log 2>&1

Write the literal repo path in place of $OBSIDIAN_WIKI_REPO — cron does not load your shell env.

Step 3: Install terminal notification (optional)

Ask the user: "Do you want a terminal reminder when your wiki is stale? (y/n)" — skip this step if they say no, or if the environment is headless/VPS.

If yes, detect the user's shell and target the right rc file:

SHELL_NAME=$(basename "$SHELL")   # zsh, bash, fish, etc.
case "$SHELL_NAME" in
  zsh)  RC_FILE="$HOME/.zshrc" ;;
  bash) RC_FILE="$HOME/.bashrc" ;;
  *)    echo "Shell '$SHELL_NAME' not auto-detected. Add the source line manually to your shell rc file." ; return ;;
esac

Check if wiki-notify.sh is already sourced in that rc file. If not, append:

echo "" >> "$RC_FILE"
echo "# obsidian-wiki terminal notification" >> "$RC_FILE"
echo "source $OBSIDIAN_WIKI_REPO/scripts/wiki-notify.sh" >> "$RC_FILE"

For Fish shell, source syntax is different — provide the manual instruction:

# Add to ~/.config/fish/config.fish:
bass source $OBSIDIAN_WIKI_REPO/scripts/wiki-notify.sh
# (requires bass plugin, or copy the logic natively)

Step 4: Run the script once

bash "$OBSIDIAN_WIKI_REPO/scripts/daily-update.sh"

This initializes $STATE_DIR/.last_update so the terminal notification works immediately.

Step 5: Confirm

Tell the user:

  • The scheduler runs daily at 9 AM (launchd and the systemd timer catch up on the next login/boot if missed; plain cron does not)
  • wiki-lint health checks run on the LINT_SCHEDULE cadence (default weekly) as part of that cycle — set LINT_SCHEDULE=daily or manual in .env to change it
  • Terminal notifications appear when the wiki is >20 hours stale
  • State is stored in <global config dir>/state/<vault-id>/ (XDG-style ~/.config/obsidian-wiki by default, or the legacy ~/.obsidian-wiki if that already exists) — supports multiple vaults independently
  • They can run /daily-update anytime to force a sync
  • Logs go to /tmp/obsidian-wiki-daily.log (launchd, cron) or journalctl --user -u obsidian-wiki-daily-update (systemd)

QMD Refresh After Vault Writes

QMD is a search index, not the source of truth. If $QMD_WIKI_COLLECTION is empty or unset, skip this step. Run it only after this skill has written or rewritten vault markdown. If QMD refresh fails, do not roll back the vault changes; report the QMD status separately.

Use $QMD_CLI if set; otherwise use qmd.

${QMD_CLI:-qmd} update

If the output says vectors are needed or embeddings may be stale, run:

${QMD_CLI:-qmd} embed

Verify the collection with either:

${QMD_CLI:-qmd} ls "$QMD_WIKI_COLLECTION"

or, when a specific page path is known:

${QMD_CLI:-qmd} get "qmd://$QMD_WIKI_COLLECTION/<page>.md" -l 5

Record one of:

  • QMD refreshed: update + embed + verified
  • QMD refreshed: update only + verified
  • QMD skipped: QMD_WIKI_COLLECTION unset
  • QMD skipped: qmd CLI unavailable
  • QMD failed: <short error summary>

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.