Stats
Actions
Tags
From osdu
Generate daily OSDU briefing notes by aggregating GitLab MRs, SPI fork health, vault goals, and brain knowledge into an insightful daily note. Use when the user says "gm", "good morning", "briefing", "daily briefing", "morning standup", "what's on my plate", or "start my day". Not for: ad-hoc status checks or single-service queries.
How this skill is triggered — by the user, by Claude, or both
Slash command
/osdu:briefingThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Generate a daily briefing note that aggregates live platform data with the brain's accumulated knowledge to produce an insightful, context-rich morning briefing.
Generate a daily briefing note that aggregates live platform data with the brain's accumulated knowledge to produce an insightful, context-rich morning briefing.
Check if the brain vault exists:
test -d "${OSDU_BRAIN:-$HOME/.osdu-brain}/00-inbox" && echo "vault: exists" || echo "vault: not found"
Then run the briefing script (always without --dry-run so the note is persisted):
uv run skills/briefing/scripts/daily-briefing.py
If the vault does not exist, the script falls back to stdout and suggests running init brain.
The vault location is $OSDU_BRAIN if set, otherwise ~/.osdu-brain (same fallback the script uses).
The briefing operates in two modes based on vault presence:
The script gathers structured data (MRs, goals, projects) and writes it to $VAULT/00-inbox/YYYY-MM-DD.md. But the script output is only the foundation. Your job as the presenting agent is to enrich it with brain knowledge (see Synthesis Protocol below).
# Generate today's briefing (writes to vault)
uv run skills/briefing/scripts/daily-briefing.py
# Preview without writing to vault
uv run skills/briefing/scripts/daily-briefing.py --dry-run
# Generate for a specific date
uv run skills/briefing/scripts/daily-briefing.py --date 2026-02-15
# Skip SPI fork health collection (faster)
uv run skills/briefing/scripts/daily-briefing.py --skip-spi
| Source | Tool | What it gathers |
|---|---|---|
| GitLab upstream | osdu-activity CLI | Your MRs, all open MRs, pipeline status |
| SPI forks | gh CLI | Fork health: issues, PRs, sync status, workflow conclusions across 8 repos |
| Obsidian vault | File system | Goals (01-goals/), projects (02-projects/) |
| Brain knowledge | QMD or file scan | Reports, RCAs, decisions, architecture notes, strategy docs |
| Azure environment | azd / az CLI | CIMPL cluster health |
| GitHub | gh CLI | Issues assigned to you |
Writes a daily note to $VAULT/00-inbox/YYYY-MM-DD.md containing:
This is what separates a mechanical status dump from an insightful briefing. After the script runs and writes the daily note, you enrich the presentation by reasoning over the brain's knowledge.
If QMD is available and its model is already downloaded (which qmd && test -d ~/.cache/qmd/models && ls ~/.cache/qmd/models/*.gguf 2>/dev/null | head -1), use it for fast, targeted searches:
# Search for content related to today's MRs, goals, or projects
qmd query "pipeline failures partition service"
qmd query "venus milestone planning"
qmd query "dependency remediation CVE"
If QMD is not available or model not yet downloaded, skip qmd and scan the vault directly:
# Recent reports (last 30 days)
find ${OSDU_BRAIN:-~/.osdu-brain}/04-reports -name "*.md" -mtime -30
# Knowledge and decisions
ls ${OSDU_BRAIN:-~/.osdu-brain}/03-knowledge/decisions/
ls ${OSDU_BRAIN:-~/.osdu-brain}/03-knowledge/*/
For each active area in today's briefing, look for relevant brain context:
| Today's data | Brain knowledge to search for |
|---|---|
| Failing MR on service X | Past RCAs for that service (04-reports/rca/), known issues |
| SPI fork alerts | Fork management RCAs, cascade history, upstream sync patterns |
| Goal at 0% progress | Related project notes, blockers, strategy docs |
| CIMPL environment status | Only use azd env list output — do not pull environment names or status from brain vault notes |
| Dependency MRs | Dependency reports, CVE coordination plans |
| Venus-related activity | Venus strategy, milestone planning, gap analysis |
| Recent MRs from others | Any organizational context the brain surfaces (initiatives, strategy) |
The value is in connecting dots the user might miss. Examples of good synthesis:
rca-master-search-azure-test) that documented a similar test environment issue. That root cause may still apply."azd env list for current deployments. The goal tracks parity between your cimpl-azure-provisioning and upstream CIMPL, not upstream environment internals."Bad synthesis (avoid):
azd env list. Never reference upstream environment names, URLs, or audit findings as if they describe your own deployment.After the script runs and you've gathered brain context, present the briefing in the terminal. The structure below is a guide, not a rigid template — adapt based on what's actually relevant today.
Good morning. It's [Day], [Date].
"[Quote]"
— [Attribution]
SPI FORKS (GitHub)
UPSTREAM SYNC — N forks have pending upstream changes
service #nn title — ⚠️ conflicts / ✅ validated / ready to merge
[clickable links to issues and PRs]
RELEASES — N pending
[clickable links: service #nn, ...]
DEPENDENCIES — N PRs across M forks
dep-name → svc1, svc2 (N forks)
[top 6 by cross-fork impact, grouped by dependency artifact]
WORKFLOWS — status (only shown if failures exist)
OSDU PLATFORM (GitLab upstream)
YOUR MRs:
> [!nnn](url) [service] — [Role] / [pipeline status] (Xd old)
RECENT (last 21 days):
> [!nnn](url) [service] @author — [pipeline status] (date) [annotation if known]
X open across Y services · Z failing
GOALS
[goal]: [progress] — [1-line brain-informed context]
PROJECTS
[project]: [phase/status] — [brain-informed insight if relevant]
BRAIN INSIGHTS
[2-4 observations connecting today's data to brain knowledge]
These should be actionable: "X relates to Y, which suggests Z"
WHAT I'D START
1. [Action] → [self / @osdu / @cimpl / @spi] — [why, informed by brain context]
2. [Action] → [self / @osdu / @cimpl / @spi] — [why]
End with: "Want me to kick off any of these?"
The generated daily note contains full URLs for every MR. Always use markdown links when presenting MRs, issues, or pipelines so the user can click through directly from the terminal:
[!nnn](https://community.opengroup.org/.../merge_requests/nnn)The URLs are already in the daily note output — extract them from there.
The script generates a full table of recent non-merged MRs from other contributors (last 21 days). Show this list — it gives the user a picture of what's happening across the platform, not just their own work.
When presenting recent MRs, cross-reference them against brain knowledge discovered during Step 1. If the brain search surfaced organizational context (e.g., initiative tracking, roadmaps, workstream plans), use what you found to annotate MRs where the connection is clear:
> [!562](url) rafs-ddms-services @valentin — ❌ Failed (Mar 20)
> [!416](url) os-core-lib-azure @aasryan — ✅ Passing (Mar 20) → [context if brain provided it]
> [!389](url) search-service @chen — ❌ Failed (Mar 18) → [context if brain provided it]
If multiple recent MRs cluster around the same area or initiative that the brain knows about, call that out in Brain Insights — it means that area is actively moving. Only annotate when the brain gave you clear context; leave MRs untagged when it didn't.
osdu-activity CLI installed (for GitLab data)gh CLI authenticated to GitHub (for SPI fork health; degrades gracefully if missing)uv run)SPI_ORG env var to override GitHub org (default: azure)SPI_SERVICES env var to override service list (space-separated)qmd CLI for fast brain knowledge search (falls back to file scan)AZURE_OPENAI_ENDPOINT + AZURE_API_KEY env vars for AI-generated quotesnpx claudepluginhub danielscholl/claude-osdu --plugin osduGenerates prioritized daily briefing for PMs: scans knowledge files (OKRs, launches, decisions, metrics, actions) and MCPs (Linear/Jira sprints, GitHub PRs/issues, Slack comms). /brief slash command.
Provides a concise briefing on current state — tasks, emails, handovers, and priorities — by reading a lazily-regenerated cache. Useful for status checks, morning reviews, or end-of-day wrap-ups.
Generates async standup notes from Git commits, Jira tickets, and Obsidian vault. Useful for remote teams wanting daily visibility without synchronous meetings.