loop-launch-a
Pre-flight checks and launch for agent loops. Validates branch state, session prompts, creates loop directory structure, and launches run.sh in a tmux session. Use when the user wants to start a new agent loop in the current repo.
What this skill does
> **Action skill** — Pre-flight, setup, and launch for agent loops.
## Input
$ARGUMENTS
---
Launch an agent loop in the current repo. This skill handles everything from validation to tmux launch.
## Pre-Flight Checks
Run these checks before launching. If any fail, report the issue and stop — do not launch a broken loop.
### 1. Git State
```bash
# Check for uncommitted changes
git status --porcelain
```
- If there are uncommitted changes, warn the user. Uncommitted work can be lost or cause conflicts during the loop.
- Suggest committing or stashing before launch.
### 2. Branch State
```bash
# What branch are we on?
git rev-parse --abbrev-ref HEAD
```
- Note the current branch — this becomes the base branch for the loop.
### 3. Existing Loops
```bash
# Check for active loops in this repo
ls agents/loops/ 2>/dev/null
```
- If other loops are running, warn about potential conflicts (especially if on the same branch).
- Check `~/.claude-loops/registry.jsonl` for start events without matching end events in this repo.
### 4. Tooling Setup
```bash
# Verify agents/ directory exists with required files
ls agents/prompts/session.md agents/prompts/worker.md agents/prompts/reviewer.md agents/run.sh 2>/dev/null
```
- If `agents/` is not set up, run the setup script:
```bash
~/claude-tooling/agents-src/setup.sh
```
- If setup script is not available, tell the user to set up the agents directory first.
## Loop ID
Derive the loop-id from the user's input:
- If the user provided a loop-id, use it directly
- If the user described a task, derive a short kebab-case slug (e.g., "fix auth flow" → `fix-auth-flow`)
- Append timestamp: `{slug}_{YYYYMMDD_HHMM}` (e.g., `fix-auth-flow_20260315_1430`)
## Session Prompts
Check if session prompts exist for this loop:
```bash
ls agents/loops/{loop-id}/session/SESSION_WORKER.md agents/loops/{loop-id}/session/SESSION_REVIEWER.md 2>/dev/null
```
If they don't exist:
1. Ask the user if they want to write them now or have you generate them
2. If generating, invoke the `loop-author-a` skill with the task description
3. If writing manually, create the directory structure and tell them where to write:
```
agents/loops/{loop-id}/session/SESSION_WORKER.md
agents/loops/{loop-id}/session/SESSION_REVIEWER.md
```
## Launch
Once pre-flight passes and session prompts exist:
```bash
# Launch the loop
./agents/run.sh --loop-id {loop-id} [additional flags]
```
`run.sh` automatically wraps itself in a tmux session named `loop-{loop-id}`.
### Additional Flags
Ask the user or infer from context:
- `--max-iter N` — default is 5
- `--worker-spec FILE` — specialization (e.g., `planner.md`, `python.md`, `rust.md`)
- `--pr-per-iteration` — enable PR-per-iteration branching
- `--claude-effort LEVEL` — `low`, `medium` (default), or `high`
## Post-Launch Output
After launching, tell the user:
1. The tmux session name: `loop-{loop-id}`
2. How to attach: `tmux attach -t loop-{loop-id}`
3. How to detach without stopping: `Ctrl+B`, then `D`
4. How to check status: point them to the `loop-observe-a` skill
5. Where state files will live: `agents/loops/{loop-id}/state/`
6. Where git-committed history will be: `.agent-loops/{loop-id}/`
## Common Patterns
| User says | What to do |
|-----------|-----------|
| "Start a loop for X" | Full pre-flight → generate prompts → launch |
| "Launch loop {id}" | Pre-flight → check prompts exist → launch |
| "Resume loop {id}" | Check if loop dir exists, session prompts present → launch with existing dir |
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.