Install
$ agentstack add skill-theveller-claude-skills-start-cc-lv ✓ 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
start-cc+lv — Claude Code + Lovable session contract
Activate this workflow at the start of any session where the project's deploy target is Lovable + Cloud Native + Render and the user previews live from origin/main.
Step 1 — Sync local to origin/main
Before any work, align the local working tree with the remote:
git status
- If working tree is clean and on
main:
``bash git fetch origin git reset --hard origin/main ``
- If working tree is dirty or on another branch: STOP. Ask the user what to do (commit + push, stash, or discard) — never
reset --hardover their work without explicit confirmation. See "Executing actions with care" in the root CLAUDE.md. - If not on
main: ask the user — do not auto-checkout; they may have intentional state.
Report the resulting HEAD sha and commit subject in one line so the user can confirm Lovable is on the same commit.
Step 2 — Session contract (active until session ends)
Once synced, follow these rules for every change made during the session:
- Single branch:
main. Never create feature branches. Lovable watchesmain. - Commit + push after every meaningful change. Pattern:
``bash git pull --rebase origin main # pick up edits made in Lovable directly git add # never -A / . — avoid pulling secrets or junk git commit -m "" git push origin main ``
- Small commits. The user rolls back via git history when something breaks — granular commits make rollback cheap.
- No destructive shortcuts without explicit user confirmation:
- No
--force/--force-with-leasepush tomain. - No
--no-verifyto skip hooks. If a hook fails, fix the root cause. - No
git reset --hardmid-session.
- Hook failures = fix the code, not bypass the hook. Re-stage and commit again (new commit, not
--amendif the original was already pushed). - Conflicts with Lovable edits: rebase, resolve, push. Do not discard remote changes — Lovable may have edited from the UI side.
Step 3 — Routine reminders to surface
- After each push, mention briefly: "Pushed `` — Lovable / Render should pick this up on next deploy."
- If push fails because remote is ahead →
git pull --rebase origin mainthen retry. Do not force. - At end of session, confirm
git statusclean andgit log @{u}..empty.
What this skill does NOT do
- Does not configure Render / Cloud Native / Lovable themselves — that wiring already exists at the GitHub repo level.
- Does not run tests or builds automatically — the user wants speed; let CI on the deploy target catch issues.
- Does not write to CLAUDE.md or memory — the skill is the source of truth for this workflow.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: TheVeller
- Source: TheVeller/claude-skills
- 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.