Install
$ agentstack add skill-manastalukdar-ai-devstudio-briefing ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Briefing
I compile a structured daily briefing from git history, open issues, TODOs, and session state so you have full situational awareness before writing a single line of code.
Token Optimization
Expected range: 300–800 tokens (all bash queries), 100 tokens (quiet day with no open items)
Patterns used: Bash for all data collection, progressive disclosure (summary → details on request), early exit per section (skip empty sections)
Early exit per section: If a section has no items (no failing tests, no open TODOs, etc.), omit it from the briefing entirely.
Step 1 — Collect Data (All Bash)
Run all queries in parallel where possible:
# Recent activity (last 24h)
git log --oneline --since="24 hours ago" 2>/dev/null | head -10
# Current branch and status
git branch --show-current
git status --short | head -20
# Open TODOs in recently changed files
git diff HEAD~3 --name-only 2>/dev/null | xargs grep -l "TODO\|FIXME\|HACK" 2>/dev/null | head -5
# Test status (if test runner available)
if [ -f "package.json" ]; then
grep -q '"test"' package.json && echo "npm test available"
fi
# Session state (if session management is active)
ls .claude/sessions/*.json 2>/dev/null | tail -1 | xargs cat 2>/dev/null
# Skill count vs declared
actual=$(ls -d skills/*/ 2>/dev/null | wc -l | tr -d ' ')
declared=$(grep -oP '(?/dev/null | head -1)
# USER.md / PROJECT.md presence
ls -la USER.md PROJECT.md 2>/dev/null
Step 2 — Assemble the Briefing
Format output as a scannable briefing card. Omit any section with no items.
── Daily Briefing ───────────────────────────────
Branch:
Recent commits (last 24h):
...
(none — clean day so far)
Status:
modified
untracked
Open TODOs in active files:
src/auth.py:42 TODO: handle token expiry edge case
...
Session context:
Last session: —
In progress:
Project health:
Skills: (declared: ) ✓ consistent / ✗ drift detected
Onboarding context:
USER.md: present / MISSING — run /project-onboard
PROJECT.md: present / MISSING
───────────────────────────────────────────────────────────────
Ready. What are we working on today?
Step 3 — Surface One Priority
After the briefing card, suggest one high-priority item based on the data:
- If tests are failing → "Suggest starting with the failing tests"
- If skill count drifts → "Suggest running /docs-sync"
- If active session exists → "Suggest resuming /session-resume "
- If no USER.md → "Suggest running /project-onboard first"
- Otherwise → "No blocking items — ready for new work"
Edge Cases
- No git repo: Skip git sections; note at the top of the briefing
- No session files: Omit session context section
- Large commit history: Cap at 10 commits regardless; offer
/session-listfor full history - CI integration: If
.github/workflows/exists andghCLI is available, include last CI run status; otherwise skip
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: manastalukdar
- Source: manastalukdar/ai-devstudio
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.