cursor-agent-supervisor
Offloading tasks with a well-defined scope to sub-agents, for instance to use a sub-agent to implement a set of specs. Use this skill whenever a task should not need a broad knowledge of the whole project
What this skill does
# Agent Supervisor You can start subagents (e.g. to work on a specific JJ revision) with: ```bash # Create a conversation with a sub-agent: cursor-agent --print --model <model-name> create-chat # Prints a conversation uuid # Give a task to the sub-agent: (command will finish when sub-agent is done) cursor-agent --print --resume <conversation-uuid> "...description of the subagent's task..." ``` ## Model Selection `cursor-agent --print --model unknown-model` will print an error that will list available models. Unless you know which model is best for the task, just use `sonnet-4.5` by default. ## Invocation **Important:** Sub-agent tasks can take several minutes. Always use a longer timeout: ```bash # In your Shell tool call, set timeout to 10 minutes timeout: 600000 ``` The `--print` flag makes the sub-agent run on its own, and reply only once it is done. ## Setting Things up for the Sub-agent Prepare things up so the agent can focus on its task (eg. if using JJ, don't ask them to run jj commands unless really necessary. Notably run `jj edit` yourself first to get into the revision where the work should be done). Ideally, the sub-agent should only have to read your instructions, hack on code, run build/tests, hack on code, etc. until your instructions are implemented, and then reply with a final answer. YOU are in charge of bookkeeping, not them. YOU have the big picture, they don't. ## Giving Good Instructions Give the subagent the instructions they will need to complete the task, but do not overwhelm them. **Do:** - Provide a clear, specific goal - List key files to read/modify - Specify what "done" looks like - Include relevant patterns to follow or reference implementations - Tell them which skills to load if needed **Don't:** - Dump or link to entire skills when they only need a subset - Include irrelevant context - Leave success criteria ambiguous ### Task Description Template ``` Work on [specific task] in [repo/directory]. **Setup:** - Skills to load or files to read IN FULL - Summarized instructions from skills or files, tailored to the task **Goal:** [Clear description of what to implement/fix] **Key files:** - path/to/main/file.ts - [why it matters] - path/to/reference.ts - [pattern to follow] **Requirements:** - [Specific requirement 1] - [Specific requirement 2] **Done when:** - [Testable criterion 1] - [Testable criterion 2] - [e.g., "pnpm -F @pkg lint passes"] ``` ## After Sub-agent Completes Always verify the sub-agent's work: 1. **Check what changed:** check the commit/revision diff, and adequation with the task 2. **Review the actual code** if the changes are non-trivial 3. **Run verification commands** (tests, type-check, lint) 4. **Update task status** based on results (if using jj-todo-workflow) ## When Things Go Wrong If the sub-agent: - **Fails or errors out:** Read the output, fix any blocking issues, retry with more context - **Produces incomplete work:** Continue the work yourself or spawn another sub-agent with clarified instructions - **Goes off track:** Review what happened, potentially revert changes, retry with tighter constraints - **Gets stuck on tooling:** May need to provide explicit commands or paths
Related in AI Agents
skill-development
IncludedComprehensive meta-skill for creating, managing, validating, auditing, and distributing Claude Code skills and slash commands (unified in v2.1.3+). Provides skill templates, creation workflows, validation patterns, audit checklists, naming conventions, YAML frontmatter guidance, progressive disclosure examples, and best practices lookup. Use when creating new skills, validating existing skills, auditing skill quality, understanding skill architecture, needing skill templates, learning about YAML frontmatter requirements, progressive disclosure patterns, tool restrictions (allowed-tools), skill composition, skill naming conventions, troubleshooting skill activation issues, creating custom slash commands, configuring command frontmatter, using command arguments ($ARGUMENTS, $1, $2), bash execution in commands, file references in commands, command namespacing, plugin commands, MCP slash commands, Skill tool configuration, or deciding between skills vs slash commands. Delegates to docs-management skill for official documentation.
reprompter
IncludedTransform messy prompts into well-structured, effective prompts — single or multi-agent. Use when: "reprompt", "reprompt this", "clean up this prompt", "structure my prompt", rough text needing XML tags and best practices, "reprompter teams", "repromptception", "run with quality", "smart run", "smart agents", multi-agent tasks, audits, parallel work, anything going to agent teams. Don't use when: simple Q&A, pure chat, immediate execution-only tasks. See "Don't Use When" section for details. Outputs: Structured XML/Markdown prompt, quality score (before/after), optional team brief + per-agent sub-prompts, agent team output files. Success criteria: Single mode quality score ≥ 7/10; Repromptception per-agent prompt quality score 8+/10; all required sections present, actionable and specific.
adaptive-compaction
IncludedAdaptive add-on policy and recovery layer that decides WHEN to compact, prune, snapshot, or fork -- replacing fixed-percent auto-compaction across Claude Code, Codex, and MCP-capable hosts. Trigger on auto-compact timing or damage: "when should I compact", "is it safe to compact now or start a fresh session", "auto-compact fires too early/mid-task", "switching to an unrelated task but the window still has space", "context rot", "answers get worse the longer the session runs", "the agent forgot the plan or my decisions after it summarized", "add a layer on top that manages context without changing the agent", raising autoCompactWindow to give the policy room, or installing/tuning a cross-tool compaction policy or PreCompact hook -- even when "compaction" is never said but the problem is context-window pressure or post-summarization memory loss. Do NOT use to summarize a conversation, build RAG, write a summarization prompt (decides WHEN not HOW), or answer max-context-length trivia.
agent-skill-creator
IncludedCreate cross-platform agent skills from workflow descriptions. Activates when users ask to create an agent, automate a repetitive workflow, create a custom skill, or need advanced agent creation. Triggers on phrases like create agent for, automate workflow, create skill for, every day I have to, daily I need to, turn process into agent, need to automate, create a cross-platform skill, validate this skill, export this skill, migrate this skill. Supports single skills, multi-agent suites, transcript processing, template-based creation, interactive configuration, cross-platform export, and spec validation.
llm-wiki
IncludedUse when building or maintaining a persistent personal knowledge base (second brain) in Obsidian where an LLM incrementally ingests sources, updates entity/concept pages, maintains cross-references, and keeps a synthesis current. Triggers include "second brain", "Obsidian wiki", "personal knowledge management", "ingest this paper/article/book", "build a research wiki", "compound knowledge", "Memex", or whenever the user wants knowledge to accumulate across sessions instead of being re-derived by RAG on every query.
skill-master
IncludedAgent Skills authoring, evaluation, and optimization. Create, edit, validate, benchmark, and improve skills following the agentskills.io specification. Use when designing SKILL.md files, structuring skill folders (references, scripts, assets), ingesting external documentation into skills, running trigger evals, benchmarking skill quality, optimizing descriptions, or performing blind A/B comparisons. Keywords: agentskills.io, SKILL.md, skill authoring, eval, benchmark, trigger optimization.