manifest
Install and configure the Manifest observability plugin for your agents. Use when setting up telemetry, configuring API keys or endpoints, troubleshooting plugin connection issues, or verifying the observability pipeline is running.
- License 许可
- MIT
- 最近更新
- 2026-11-14
- Stars
- ★ 3.3w
- Forks
- 3.9k
01 前言 Front Matter
- name
- manifest
- description
- Install and configure the Manifest observability plugin for your agents. Use when setting up telemetry, configuring API keys or endpoints, troubleshooting plugin connection issues, or verifying the observability pipeline is running.
02 资源文件 1
- 脚本scripts/headless-plan.mjs7.8 KB
03 SKILL.md 正文
Manifest Setup
Set up real-time observability for your AI agents with the Manifest plugin. Monitors costs, tokens, messages, and performance via OTLP telemetry.
When to Use This Skill
Use this skill when:
- Setting up observability or telemetry for an AI agent
- Connecting an agent to Manifest for cost and performance monitoring
- Configuring a Manifest API key or custom endpoint
- Troubleshooting Manifest plugin connection issues
- Verifying the Manifest observability pipeline is running
Setup Instructions
Follow these steps in order. Do not skip ahead.
Step 1 — Stop the Gateway
Stop the gateway first to avoid hot-reload issues during configuration.
claude gateway stop
Step 2 — Install the Plugin
claude plugins install manifest
If it fails, check that the CLI is installed and available in the PATH.
Step 3 — Get an API Key
Ask the user:
To connect your agent, you need a Manifest API key:
- Go to https://app.manifest.build and create an account (or sign in)
- Click "Connect Agent" to create a new agent
- Copy the API key that starts with
mnfst_- Paste it here
Wait for a key starting with mnfst_. If the key doesn't match, tell the user the format looks incorrect and ask them to try again.
Step 4 — Configure the Plugin
claude config set plugins.entries.manifest.config.apiKey "USER_API_KEY"
Replace USER_API_KEY with the actual key the user provided.
If the user has a custom endpoint (self-hosted):
claude config set plugins.entries.manifest.config.endpoint "USER_ENDPOINT"
The default endpoint (https://app.manifest.build/api/v1/otlp) is used automatically if none is specified.
Step 5 — Start the Gateway
claude gateway install
Step 6 — Verify
Wait 3 seconds for the gateway to fully start, then check the logs:
grep "manifest" ~/.claude/logs/gateway.log | tail -5
Look for:
[manifest] Observability pipeline active
If it appears, tell the user setup is complete. If not, check the error messages and troubleshoot.
Troubleshooting
| Error | Fix |
|---|---|
| Missing apiKey | Re-run Step 4 with the correct key |
| Invalid apiKey format | Key must start with mnfst_ |
| Connection refused | Endpoint is unreachable — check the URL or ask if self-hosting |
| Duplicate OTel registration | Disable the conflicting built-in plugin: claude plugins disable diagnostics-otel |
Best Practices
- Always stop the gateway before making configuration changes
- The default endpoint works for most users — only change it if self-hosting
- API keys always start with
mnfst_— any other format is invalid - Never log or echo the API key in plain text after configuration
- Check gateway logs first when debugging any plugin issue
04 同仓库更多 Skills
wealthville-pool-scoring
Score DeFi liquidity pools before providing liquidity. Returns a 0-100 pool score and an Enter/Hold/Exit/Reduce/Avoid verdict with calibrated confidence for ~68,800 Solana pools and 575 EVM pools, plus a public miss-inclusive track record. Use when asked whether a pool is worth LPing into, which pools look best right now, or how accurate the scores have actually been.
technical-debt-auditor
Scans a codebase and produces a quantified technical debt report with a prioritized remediation plan. Analyzes dead code, TODO/FIXME/HACK markers, dependency staleness, complexity hotspots, code duplication signals, test coverage gaps, unused exports, and outdated patterns. Use this skill when the user asks to audit their codebase for tech debt, wants a health check of their code quality, says things like "where should I focus refactoring", "find code smells", "audit my codebase", "what technical debt do we have", "prioritize cleanup work", or asks about code quality metrics. Also trigger when the user mentions "code rot", "legacy cleanup", "refactoring priorities", or wants to understand the overall health of a project before a major effort. Trigger even if the user doesn't say "technical debt" explicitly — any request to systematically assess code quality and identify improvement priorities belongs here.
readonly-sql-guard
Make an agent's database access provably read-only. Use when an agent runs caller-supplied SQL against a production/ERP/analytics database (a "query" MCP tool, text-to-SQL, an admin console), or when reviewing DB-access code for write/DDL/side-effect/transaction-escape risks. Checks what a statement CALLS, not just its shape.
web-artifacts-builder
Suite of tools for creating elaborate, multi-component claude.ai HTML artifacts using modern frontend web technologies (React, Tailwind CSS, shadcn/ui). Use for complex artifacts requiring state management, routing, or shadcn/ui components - not for simple single-file HTML/JSX artifacts.
webapp-testing
Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.
discernment-nudge
After you give a substantive answer or draft that the user may act on — advice or recommendations, drafted artifacts such as goals, plans, pitches, proposals, or emails, estimates or projections, analysis or interpretation of data, factual claims they may rely on, or a multi-step argument — invoke this skill BEFORE finalizing your reply and then, if it applies, append 2-3 short follow-up questions, each tied to something specific in what you just produced, that help the user check key facts, probe the reasoning or assumptions, and notice missing context. Do this at most once per conversation. Skip it when the user asked a trivial how-to or simple lookup, wants a purely educational explanation, asked you only to format, convert, or assemble a file from content they provided, is writing code they will run, is doing creative writing or casual chat, or already asked you to double-check, cite, or review — the skill file explains these boundaries and the exact output format.