co8/manage-vercel-skills
skills/manage-vercel-skills/SKILL.md
Manage which Vercel plugin skills are enabled or disabled. Use when the user wants to disable noisy Vercel skills, enable previously disabled skills, or see which skills are active. Invoked via /manage-vercel-skills.
$ install --global
skillsauthnpx skillsauth add co8/cc-plugins manage-vercel-skillsInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
Security Scan Results
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
3
Scanners Passed
9
Scanners in report
Clean
TrivyContainer and dependency vulnerability scanner
95%
Clean
SemgrepStatic code analysis for vulnerabilities
95%
Clean
mcp-scan (Snyk)Model Context Protocol security validation
95%
Skipped
Snyk (dep)Open source security scanning
50%
Skipped
Socket.devSupply chain security analysis
50%
Skipped
VirusTotalMulti-engine malware detection
50%
Skipped
CrowdStrikeAdvanced threat intelligence
50%
Skipped
OSV-ScannerOpen Source Vulnerability database check
50%
Skipped
OWASP Dep-Check
50%
Last scanned: May 13, 2026, 2:44 AM161.2s1 file scanned
SKILL.md
- name:
- manage-vercel-skills
- version:
- 1.0.0
- description:
- Manage which Vercel plugin skills are enabled or disabled. Use when the user wants to disable noisy Vercel skills, enable previously disabled skills, or see which skills are active. Invoked via /manage-vercel-skills.
Manage Vercel Skills
Disable or enable individual skills from the Vercel plugin by managing the VERCEL_PLUGIN_SEEN_SKILLS env var in ~/.claude/settings.json.
How it works
The Vercel plugin uses VERCEL_PLUGIN_SEEN_SKILLS (comma-separated list) to track which skills have already been injected in a session. Skills in this list are skipped by the dedup system. By pre-populating this env var in settings.json, skills are permanently suppressed across all sessions.
Instructions
When this skill is invoked, follow these steps:
1. Read current state
Read ~/.claude/settings.json and extract the env.VERCEL_PLUGIN_SEEN_SKILLS value. Parse it as a comma-separated list of disabled skill names.
2. Get available skills
List all directories in the Vercel plugin skills folder:
~/.claude/plugins/cache/claude-plugins-official/vercel/*/skills/
Skip entries starting with _ (internal files like _chain-audit.md).
3. Show the user a table
Display a table with ALL Vercel skills showing their status:
| # | Skill | Status |
|---|-------|--------|
| 1 | agent-browser | enabled |
| 2 | ai-sdk | DISABLED |
...
4. Ask what to change
Ask the user which skills to enable or disable. Accept:
- Skill names:
disable ai-sdk ai-elements chat-sdk - Numbers from the table:
disable 3,5,8 - Bulk:
disable all,enable all - Toggle:
toggle ai-sdk
5. Update settings.json
Edit ~/.claude/settings.json to update the VERCEL_PLUGIN_SEEN_SKILLS value in the env object:
- If the
envkey doesn't exist, create it - If
VERCEL_PLUGIN_SEEN_SKILLSdoesn't exist, create it - Write the comma-separated list of disabled skill names (sorted alphabetically)
- If no skills are disabled, remove the key entirely
6. Confirm
Show the updated table and remind the user the changes take effect on the next session (or after /clear).
Important notes
- This only works for the Vercel plugin. Other plugins don't use
VERCEL_PLUGIN_SEEN_SKILLS. - Plugin updates may add new skills — they'll be enabled by default until explicitly disabled here.
- On context compaction, the Vercel plugin may re-inject high-priority skills even if they're in the seen list. This is by design for critical skills during long sessions.
Related Skills
co8/council-review
development
VerifiedTrustedCommunity
Multi-agent code review council. A parallel team of specialized read-only reviewers (architect, correctness, security, compliance, silent-failures, type/comment, simplification) audits a diff, a synthesis agent consolidates findings, and a debate-consensus pass resolves conflicts into a single ranked verdict. Use this skill when reviewing a PR, branch, or diff before merge — standalone, or as the review primitive invoked by swarm-protocol's review-repair phase. Triggers on: (1) `/council-review <PR URL | branch | diff>`, (2) `/council-review` (no args) to review current uncommitted changes, (3) requests to review code with multiple specialized lenses or to gate a merge.
1SKILL.mdUpdated May 27, 2026
co8/swarm-protocol
development
VerifiedTrustedCommunity
Multi-agent development orchestration for complex projects. Use this skill when orchestrating parallel development workstreams, coordinating multiple agent tasks, managing project documentation structure, or executing `/swarm-protocol` commands. Triggers on: (1) `/swarm-protocol <project-name>` to initialize new projects with full planning, (2) `/swarm-protocol` (no args) to continue existing or start new project, (3) requests involving parallel agent coordination, milestone commits, or multi-phase development workflows.
1SKILL.mdUpdated May 13, 2026
co8/secret-sauce
tools
VerifiedTrustedCommunity
Development best practices and project patterns. Use when starting projects, setting up CLAUDE.md, coding TypeScript/Next.js/React/Supabase, implementing AI flows, data fetching, testing, deployment, git workflows, browser automation, centralized configuration, or Tailwind CSS v4.
1SKILL.mdUpdated May 13, 2026
openclaw/taskflow
tools
VerifiedTrustedCommunity
Use when work should span one or more detached tasks but still behave like one job with a single owner context. TaskFlow is the durable flow substrate under authoring layers like Lobster, ACPX, plugins, or plain code. Keep conditional logic in the caller; use TaskFlow for flow identity, child-task linkage, waiting state, revision-checked mutations, and user-facing emergence.
357,764SKILL.mdUpdated Apr 10, 2026