Name: agent-launcher-orchestrator
Rating: 76.8 (1 reviews)
Author: alirezarezvani

agent-launcher-orchestrator

Use when a user wants to build, launch, grade, or schedule a Claude Managed Agent (CMA) in their own Anthropic account — "build me an agent", "launch this as a managed agent", "run this on a schedule", "grade my agent against a rubric", "set up a nightly worker". Reads the per-session goal (./my-agent/goal.json), routes deterministically to one of five phase sub-skills (interview → stage-launch → grade-iterate → run-without-you → wrap-up) via goal_router.py, and compiles the goal+phase into an execution shape (single-pass workflow / bounded grade→iterate loop / recurring cron deployment loop) via loop_compiler.py. Forks context so heavy intake (build sheets, payloads, eval cases) stays out of the parent thread. All launches are emitted as BYOK curl the user runs with their own key; no tool makes API calls. Inspired by anthropics/launch-your-agent (Apache-2.0). Distinct from engineering/agent-harness (generic domain loop) and engineering/write-a-skill (authors Claude Code skills, not CMAs).

Quality

98%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

The canonical home for this skill is agent-launcher-orchestrator in alirezarezvani/claude-skills

Quality

Content

96%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is an exceptionally lean, actionable orchestrator spec with concrete commands, clear sequencing, and validation checkpoints. Its only gap is progressive disclosure: the markdown links and scripts it points to are not bundled alongside SKILL.md.

Suggestions

Bundle the referenced files so the links resolve: add references/cma-primitives.md and references/loops-and-workflows.md (and interview-to-config.md) under a references/ directory, or adjust the link paths to match the actual bundle layout.

Include the three scripts the body instructs Claude to run (scripts/goal_state.py, scripts/goal_router.py, scripts/loop_compiler.py) in a scripts/ directory so the executable commands are runnable as written.

Verify the ../../references/ relative paths against the plugin's install location; if the references live elsewhere, note that in the body so Claude does not search for missing files.

Dimension	Reasoning	Score
Conciseness	The ~50-line body is dense and lean — routing table, compile command with concrete flags, pre-flight gates, hand-off contract, and forcing-question library each earn their place with no padding or explanation of concepts Claude already knows.	5 / 5
Actionability	Provides copy-paste-ready executable commands ("python3 scripts/goal_router.py --out-dir ./my-agent" with documented exit codes, a full loop_compiler.py invocation with --max-iterations/--cron/--timezone/--nest-outcome flags) plus Recommend/Cite guidance per forcing question.	5 / 5
Workflow Clarity	Sequences the multi-step flow explicitly (router → act on exit code → compile loop → fork to sub-skill → advance → digest) with validation checkpoints and recovery loops (exit 3 ASK → ask → re-route; exit 4 REFUSE → get one sentence → re-route) and hard pre-flight refusals.	5 / 5
Progressive Disclosure	The body is a well-sectioned overview with clearly signaled one-level-deep references (cma-primitives.md, loops-and-workflows.md, interview-to-config.md, named sub-skills), but the referenced bundle files (references/.md and scripts/.py) are not present in the skill's bundle, a minor organization gap.	4 / 5
	Total	19 / 20 Passed

Description

100%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is excellent: it pairs an explicit 'Use when...' trigger with concrete natural user phrases, names specific scripts and execution shapes, and explicitly differentiates the skill from siblings. Both what and when are answered with minimal fluff.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions with named scripts — "routes deterministically to one of five phase sub-skills", "compiles the goal+phase into an execution shape ... via loop_compiler.py", "Forks context", and "emitted as BYOK curl" — giving comprehensive coverage rather than vague abstraction.	5 / 5
Completeness	Explicitly answers both what and when with concrete triggers — "Use when a user wants to build, launch, grade, or schedule a Claude Managed Agent (CMA)" followed by verbatim example user phrases.	5 / 5
Trigger Term Quality	Includes natural phrases a user would actually say ("build me an agent", "launch this as a managed agent", "run this on a schedule", "grade my agent against a rubric", "set up a nightly worker") with strong synonym coverage across build/launch/grade/schedule.	5 / 5
Distinctiveness Conflict Risk	States a clear niche (CMAs in the user's own Anthropic account, BYOK curl, no API calls) and explicitly distinguishes itself: "Distinct from engineering/agent-harness ... and engineering/write-a-skill", minimizing conflict risk.	5 / 5
	Total	20 / 20 Passed

Validation

81%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 2 suspicious	Warning
referenced_paths_exist	Referenced path issues: 5 missing	Warning

	Total	13 / 16 Passed

Repository: alirezarezvani/claude-skills
Path: .gemini/skills/agent-launcher-orchestrator/SKILL.md
Commit: 19392f7

Reviewed: ~1 hour ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.