Manager
Use this agent when the user requests an action that should trigger 'documentation' behavior. This agent is usually triggered by the obsidian-project-documentation skill.
- Type
- Subagent
- Repository
- jeremylongshore/tons-of-skills-marketplace
- GitHub stars
- 2.8k
- License
- MIT
- Repo last updated
- Sep 27, 2026
- Model
- sonnet
What Manager is
Manager is a subagent published in the jeremylongshore/tons-of-skills-marketplace repository on GitHub, which has about 2.8k stars. The repository describes itself as: “Model-agnostic agent-skills platform with a harness-free canonical layer, verified adapters, and the ccpi package manager. Explore at tonsofskills.com.”
A subagent is a specialist assistant that Claude can hand part of a task to. It is a markdown file whose frontmatter sets a name, a description that tells Claude when to delegate, and optionally the tools and model it may use; the body becomes the subagent's own system prompt.
Because a subagent works in its own context, it keeps the main conversation focused: Claude can send a narrow job, such as a review or a specialised analysis, to Manager and get back a compact result.
It is set up to use these tools: Read, Write, Edit, Bash, TodoWrite, AskUserQuestion, Glob, Grep. Limiting tools is a good sign: the subagent can only do what those tools allow.
How to install Manager
Claude Code
- Download manager.md from the repository.
- Save it to ~/.claude/agents/ to use it in every project, or to .claude/agents/ inside one project to share it through version control.
- Claude Code watches these folders, so the subagent is usually available right away. Ask Claude to use it by name, or @-mention it to make sure it runs.
Claude Cowork
- Cowork loads subagents through plugins. If the repository is packaged as a plugin marketplace, add it under Customize → Plugins → Add marketplace and install the plugin that contains this subagent.
- Otherwise, bundle the file into your own plugin's agents/ folder and upload it from Customize → Plugins.
New to extending Cowork? Our plugins guide and Customize guide explain how skills, plugins, and connectors fit together.
Inside the source file
An excerpt from plugins/productivity/obsidian-project-documentation/agents/manager.md, shared under the repository's MIT license. Read the full file on GitHub.
You are the Obsidian Project Documentation Manager agent. You are a meticulous project documentation manager specializing in technical documentation workflows for the User's projects. Your expertise lies in capturing decisions, maintaining project continuity, ensuring seamless handoffs between work sessions, and keeping all documentation in a consistent structure within the User's Obsidian vault.
Context you need
- Vault Path: {vault_path}
- Project Name: {project_name}
- Project Area: {area}
- Description: {description}
- Working Directory: {cwd}
- Current Date: {current_date}
- The Claude Code session conversation and skill that trigger you
Configuration you refer to
- auto_commit: {auto_commit}
- auto_push: {auto_push}
- git_enabled: {git_enabled}
User's request you refer to
{user_original_message}
Special handling for meta-documentation
If the working directory ({cwd}) contains "obsidian-project-assistant":
- You are documenting the documentation tool itself (META situation)
- ALL 8 steps (below) still apply
- Step 5 is CRITICAL: Update CLAUDE.md in {cwd} to reflect any architectural changes, new features, or refactoring
- Both the Obsidian vault note AND the repository's own documentation must be updated
Your tasks
CRITICAL: Before starting work, use the TodoWrite tool to create a task list with all 8 steps below. Mark each step as "in_progress" when you begin it and "completed" when finished. This ensures nothing is skipped and provides visibility to the user.
When activated, you will:
1. Create or update the appropriate Obsidian note based on the User's request
- Check if note exists at: {vault_path}/Projects/{project_name}.md
- If the Projects folder in the User's Obsidian vault doesn't exist, create it.
- If new: Locate the template using:
find ~/.claude/plugins/cache/ali5ter/obsidian-project-documentation -name "project-template.md" 2>/dev/null | head -1Read the file at the returned path.
- Generate values for all placeholders before substitution:
- {{title}} — project name
- {{date}} — run date +%Y-%m-%d
- {{time}} — run date +%H:%M
- {{area}} — detected or user-supplied area
- {{area_tag}} — area converted to lowercase with hyphens (e.g., "Music Synthesis" → "music-synthesis", "Hardware" → "hardware")
- {{phase}} — evaluated from the phase progression Planning → Implementing → Testing → Complete based on project state; phases can bounce between Implementing and Testing or move back when appropriate
- {{description}} — project description
- Fill all placeholders in the template with the generated values.
- If updating: Read existing note, preserve content, append another Update section with content detailed in step 3 (progress extraction) below.
- Update the 'updated:' field in frontmatter to {current_date}.
2. Analyze cross-project relationships and update relationship metadata
This step builds knowledge connections across the vault to power Obsidian's graph view.
Extract technologies from the current project:
- Scan the working session conversation, README.md, package.json, requirements.txt, Cargo.toml, go.mod, or any other dependency/config files present in {cwd}
- Identify canonical technology names by matching against the "Canonical Technology Names for Relationship Matching" section in area-mapping.md. Locate it using:
find ~/.claude/plugins/cache/ali5ter/obsidian-project-documentation -name "area-mapping.md" 2>/dev/null | head -1- Write these to the technologies: frontmatter array (e.g., technologies: [Arduino, ESP32, MQTT, I2C])
- Use canonical names exactly as listed in area-mapping.md for consistency across notes
Scan existing project notes for relationship candidates:
- Run: ls {vault_path}/Projects/*.md 2>/dev/null
- Exclude the current project file ({project_name}.md) from this list
- For each candidate file, read only the first 30 lines (frontmatter + Overview section) to keep token cost low
- Extract: area, technologies, and the first paragraph of the Overview
Score each candidate for relationship strength:
Before you install
- Read the whole file first. Skills, commands, and subagents are instructions Claude will follow, so make sure they match what you want.
- Check which tools, scripts, or MCP servers it uses. Local servers and scripts run with your permissions.
- Try it in a test project or a copy of your files before pointing it at real work.
- Pin the version you tested, and review changes before updating.
- Watch for instructions that fetch web content or run shell commands; those are where prompt injection risks start. See our prompt injection guide.
FAQ
What is Manager?
Manager is a subagent for Claude Code and Claude Cowork from the jeremylongshore/tons-of-skills-marketplace repository on GitHub. Use this agent when the user requests an action that should trigger 'documentation' behavior. This agent is usually triggered by the obsidian-project-documentation skill.
How do I install Manager in Claude Code?
Download manager.md from the repository. Save it to ~/.claude/agents/ to use it in every project, or to .claude/agents/ inside one project to share it through version control. Claude Code watches these folders, so the subagent is usually available right away. Ask Claude to use it by name, or @-mention it to make sure it runs.
Can I use Manager in Claude Cowork?
Cowork loads subagents through plugins. If the repository is packaged as a plugin marketplace, add it under Customize → Plugins → Add marketplace and install the plugin that contains this subagent. Otherwise, bundle the file into your own plugin's agents/ folder and upload it from Customize → Plugins.
Is Manager safe to install?
It is a third-party community resource, not reviewed by Anthropic or this site. Read the source file first, check which tools and connectors it uses, and install only from sources you trust.
Similar resources
- Veeva Pack Claude Code skill pack for Veeva (24 skills) Plugin · jeremylongshore/tons-of-skills-marketplace
- Vector Db Expert Selects and configures the right vector database (Pinecone, Qdrant, Weaviate, pgvector, Milvus, ChromaDB) based on scale, budget, latency, and query patterns, with HNSW tuning and migration guidance. Use when choosing or optimizing a vector store for a RAG or semantic search system. Trigger with "which vector database should I use", "optimize my vector DB". Subagent · jeremylongshore/tons-of-skills-marketplace
- Vertex Campaign Generate complete multimodal marketing campaigns using Vertex AI - video, Slash Command · jeremylongshore/tons-of-skills-marketplace
- Vibe Explainer User-facing progress translator that reads .vibe/status.json and renders structured, jargon-free summaries of what changed, what was verified, what's next, and any required user actions — with a dedicated error mode for failures. Use when technical progress needs to be communicated in plain language or when checking vibe-guide session status. Trigger with \"vibe status\", \"/vibe-guide:status\". Subagent · jeremylongshore/tons-of-skills-marketplace
- Mark Designs and stewards visual identities — logo usage rules, brand guidelines, asset libraries, and visual identity systems that stay consistent across contexts. Use when creating brand guidelines or auditing existing brand assets. Trigger with \"write brand guidelines\", \"audit our brand assets\". Subagent · jeremylongshore/tons-of-skills-marketplace
- Make Expert Design and document Make.com (Integromat) visual automation scenarios with modules, routers, error handling, and cost estimates. Use when building or architecting a Make.com integration workflow. Trigger with "design a Make scenario", "build a Make workflow". Subagent · jeremylongshore/tons-of-skills-marketplace
- Memory Agent Maintains rolling 90-day analytics baselines, learned seasonal patterns, and anomaly history so specialist agents never start cold. Use when reading prior context before a report or writing updated baselines after one. Trigger with \"load analytics context\", \"update analytics baselines\". Subagent · jeremylongshore/tons-of-skills-marketplace
- Macro Economist Translates macroeconomic data (GDP, CPI, Fed policy, yield curve) into asset-class positioning and sector rotation recommendations across business cycle phases. Use when understanding macro impact on markets or building a macro-driven portfolio view. Trigger with "macro outlook", "what does this Fed decision mean for markets". Subagent · jeremylongshore/tons-of-skills-marketplace