Convex Live Deployment Advisor
Turns Convex runtime insights (read limits, write contention) into code-backed performance and cost fixes
- Repository
- openclaw/clawhub
- Stars
- 9,546
- Licence
- MIT
- Files
- 1 file
What it does
This skill makes the agent review a live Convex backend's last 72 hours of runtime health data, such as functions hitting read limits or repeatedly retrying writes, instead of guessing from static code. For each flagged event it reads the relevant function's code, identifies the likely cause (for example an unindexed filter or a hot shared counter), and reports ranked findings with evidence and suggested fixes. It does not change anything on its own; fixes are applied only after the user confirms.
Great for
- Developers running production or cloud dev Convex apps who want to find slow or costly queries
- Teams investigating write contention or hitting Convex read limits
- People who want performance findings tied to real traffic data rather than general advice
Skip it if
- Projects without about 72 hours of traffic or on preview/deploy-key contexts, where the insights data is unavailable and the skill will report nothing rather than invent findings
- Security or authorization reviews, which the skill explicitly hands off to other skills; it is limited to performance, cost and health
- Anyone expecting it to fix code automatically without review, since it is designed to apply changes only on confirmation
Try asking
- “Check my Convex deployment's last 72 hours of insights and tell me which functions are costing the most reads”
- “Why is my messages:list function hitting the document read limit? Find the root cause in the code”
- “Several of my mutations keep retrying on the same document; find the contention source and suggest a fix”
- “Review my Convex production health and rank the performance findings by severity”
Install in 3 steps
- 1
Download the skill
The ZIP contains the 1 original files, the MIT licence and an attribution note.
convex-advisor.zip - 2
Add it to your agent
Claude Code (all projects)
mkdir -p ~/.claude/skills && unzip convex-advisor.zip -d ~/.claude/skills/Claude Code / Agent SDK (this project only)
mkdir -p .claude/skills && unzip convex-advisor.zip -d .claude/skills/Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.
- 3
Ask for the task
No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.
Get better results
- Run it on a deployment with real traffic; a fresh or idle deployment will produce no findings by design
- Ask for a fix only after reviewing the ranked findings, then re-run the insights check later to confirm the trend improved
- Pair it with the convex-reviewer skill for code style issues and convex-authz for permission checks, since this skill intentionally stays within performance and cost
What's inside
- SKILL.md
Redistributed unmodified from openclaw/clawhub under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.