Skip to content

dvy1987/agent-loom / .agents/skills/agent-launcher

agent-launcher

Verified

Internal skill. Called by setup-evaluation after a PASS. Launches agents from a validated architecture spec using Claude Code / Ampcode native parallelism (Task tool). Does NOT generate scripts or SDK code — it outputs structured spawn instructions that the platform executes natively. Never invoked directly by the user. Never launches without a setup-evaluation PASS.

FileKindSize
SKILL.mdskill md7.4 KB
references/examples.mdmarkdown1.9 KB

references/examples.md

View on GitHub
# Agent Launcher — Full Worked Examples

Skill: `agent-launcher` | Enriched from SKILL.md (improve-skills pass, SKIP_RESEARCH).

## Example 1 — Documented workflow

**Input:** Architecture: proc-2026-04-11-012 (3 parallel agents: review-agent, security-agent, debt-agent). Platform: Ampcode. setup-evaluation: PASS. All prompt files present.

**Output:**
```
Preconditions: ✓ Arch spec ✓ setup-evaluation PASS ✓ All prompts present
Platform: Ampcode (Task tool)
Manifest written: docs/agents/runs/2026-04-11-review-pipeline-manifest.md

SPAWN SUBAGENTS:
Topology: parallel

Agent: review-agent
Role prompt: docs/agents/review-agent-prompt.md
Input: docs/handoffs/task-input.md
Output to: docs/handoffs/review-agent-output.md

[repeat for each agent in architecture spec]

Run all agents concurrently via Task tool. Wait for all outputs.
```

## Example 2 — Step-by-step execution

**Input:** "Run `agent-launcher` on [concrete task]"

**Agent actions:**
1. Precondition Check
2. Platform Check
3. Read Architecture Spec
4. Write Launch Manifest
5. Spawn Instructions
6. Monitor and Hand Off

**Impact Report shape:**
```
Agents launched: [N]
Platform: Claude Code / Ampcode (Task tool native)
Topology: sequential | parallel | hierarchical
Manifest: docs/agents/runs/YYYY-MM-DD-<slug>-manifest.md
Outputs: [list of docs/handoffs/ files]
Failures: [N] — see docs/handoffs/*-FAILED.md
Next: project-orchestrator (synthesis)
```

## Example 3 — Gotcha application

**Input:** Task hits a non-obvious edge case

**Apply:**
- Parallel agents must be truly independent — if Agent B needs Agent A's
- Sequential agents must chain outputs explicitly — never assume an agent
- Prompt files must exist on disk before spawning — in-memory prompts do
- The launch manifest is the audit trail — never skip it even for two-agent runs.

---

See `SKILL.md` for hard rules and verification checklist.
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 471kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 322kUpdated Claude CodeCodexVerifiedInstall