launch-sub-agent

Launch an intelligent sub-agent with automatic model selection based on task complexity, specialized agent matching, Zero-shot CoT reasoning, and mandatory self-critique verification

Quality

64%

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

Fix and improve this skill with Tessl

tessl review fix ./plugins/sadd/skills/launch-sub-agent/SKILL.md

Quality

Content

70%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 content is highly actionable with a clear, well-validated multi-phase workflow, but it is verbose and keeps large reusable templates inline rather than offloading them to one-level-deep reference files.

Suggestions

Move the full Zero-shot CoT prefix and Self-Critique suffix templates into reference files (e.g. references/cot-prefix.md, references/critique-suffix.md) and reference them by path, leaving only the assembly instructions inline.

Trim explanatory prose inside the templates that restates reasoning steps Claude already performs unsupervised (e.g. the 'Let me first understand...' walkthrough), keeping only the mandatory structural requirements.

Add a short 'Quick start' summary at the top showing the minimal dispatch recipe, so the overview is scannable before the detailed phases.

Dimension	Reasoning	Score
Conciseness	The ~330-line body is well structured, but the Zero-shot CoT prefix (4.1) and Self-Critique suffix (4.3) over-explain reasoning and verification concepts Claude already applies, adding padded templating that could be trimmed.	3 / 5
Actionability	Provides a concrete model-selection table, an ASCII decision tree, named specialized agents, a templated sub-agent prompt with labeled placeholders, and the Task tool dispatch fields (description/prompt/model); gaps are minor since prompts still need per-task filling.	4 / 5
Workflow Clarity	A clear five-phase sequence (analyze → select model → match agent → build prompt → dispatch) with an explicit mandatory self-critique feedback loop (STOP/FIX/RE-VERIFY/DOCUMENT) and a 5-question verification checklist, matching the highest anchor.	5 / 5
Progressive Disclosure	Section structure is good, but the body is a ~330-line monolith with the large CoT and self-critique templates inlined rather than split into reference files, and no external references exist (the under-50-line simple-skill exception does not apply).	3 / 5
	Total	15 / 20 Passed

Description

58%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 conveys a clear, multi-mechanism 'what' but omits any 'when to use it' trigger guidance, which is its main weakness. It also leans on technical jargon (Zero-shot CoT, self-critique) over natural user phrasing.

Suggestions

Add an explicit 'Use when...' clause naming natural triggers, e.g. 'Use when delegating a focused task to a sub-agent, spinning up a worker for an isolated subtask, or automating a piece of multi-step work.'

Reword jargon-heavy phrases ('Zero-shot CoT reasoning', 'self-critique verification') into user-facing language like 'step-by-step reasoning' and 'built-in verification' so trigger terms match what users actually say.

Lead with the observable outcome ('Dispatch a focused sub-agent to execute a task...') before listing the internal selection mechanisms.

Dimension	Reasoning	Score
Specificity	Lists several specific mechanisms ('automatic model selection based on task complexity, specialized agent matching, Zero-shot CoT reasoning, and mandatory self-critique verification'), naming concrete capabilities beyond a single action, though they describe internal mechanisms rather than observable deliverables.	4 / 5
Completeness	It clearly states what the skill does ('Launch an intelligent sub-agent with automatic model selection...') but provides no 'Use when...' clause or equivalent explicit trigger guidance, which caps completeness at 3 per the judging guidelines.	3 / 5
Trigger Term Quality	'sub-agent' and 'model selection' are relevant keywords a user might say, but 'Zero-shot CoT reasoning' and 'self-critique verification' are technical jargon and common natural phrasings like 'delegate a task' or 'automate a task' are missing.	3 / 5
Distinctiveness Conflict Risk	It carves a clear niche in sub-agent orchestration with model/agent selection, mostly distinct from other skills with only minor overlap risk against generic delegation tooling.	4 / 5
	Total	14 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: NeoLabHQ/context-engineering-kit
Path: plugins/sadd/skills/launch-sub-agent/SKILL.md
Commit: 8539779

Reviewed: 4 days 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.