delegate-to-agent

How to delegate all AI work to the agent chat. Use when delegating AI work from UI or scripts to the agent, when a user asks for agent behavior or LLM-powered features, when tempted to add inline LLM calls, or when sending messages to the agent from application code.

Quality

80%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

82%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.

A highly actionable, well-structured skill body driven by concrete code examples and clear decision criteria for delegation. The main gaps are minor: a little trimmable explanatory prose and no explicit validation checkpoints in the decision flows.

Suggestions

Tighten the 'Why' section and de-duplicate the 'Don't' list against the 'Rule' to recover token budget.

Consider splitting the 'Delegating to a Sub-Agent' and 'When to Use A2A' sections into one-level-deep reference files so SKILL.md stays a lean overview.

Add an explicit verification checkpoint for background delegation (e.g., how to confirm a background run succeeded via run state) to strengthen workflow clarity.

Dimension	Reasoning	Score
Conciseness	The body is dense and actionable with no padding of basic concepts Claude already knows; every section ties to a concrete API or decision. It is not a 5 because the 'Why' section and a few explanatory paragraphs (e.g. restating the rule in 'Don't') could be trimmed, but it is clearly above the 3 anchor's unnecessary explanations.	4 / 5
Actionability	Fully executable, copy-paste-ready code covers the common cases — sendToAgentChat (foreground/background), completeText, useAgentChatGenerating, and callAgent — with real imports, parameters, and a submit-vs-prefill table. This matches the 5 anchor where specific examples cover the common cases.	5 / 5
Workflow Clarity	Decision flows are clearly sequenced with explicit criteria and checklists (sub-agent briefing contract's four required fields, submit/prefill table, fan-out cap of ~3). It is not a 5 because there are no explicit validation/feedback-loop checkpoints, and not a 3 because the sequence and gates are explicit rather than merely implied.	4 / 5
Progressive Disclosure	The single self-contained file is well-sectioned (Rule, Why, How, Don't, Exception, A2A, Related Skills) with no nested references and clear cross-skill links. It is not a 5 because sub-agent and A2A detail is inlined rather than split into one-level-deep reference files, and not a 3 because structure and signaling are already good.	4 / 5
	Total	17 / 20 Passed

Description

78%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.

A strong description with an explicit 'Use when...' clause and concrete, natural trigger phrases covering the delegation domain. Its main weakness is specificity — it states the approach rather than enumerating the concrete capabilities the skill governs.

Suggestions

Add 1-2 concrete capability verbs (e.g., 'send messages, prefill the composer, spawn sub-agents, and run narrow server-side text transforms via the agent') to lift specificity.

Disambiguate 'the agent' as the local agent in the description so it does not overlap with the a2a-protocol skill's triggers.

Dimension	Reasoning	Score
Specificity	Names the domain ('delegate all AI work to the agent chat') and a couple concrete actions (delegating, 'sending messages to the agent from application code'), but does not enumerate the full set of capabilities. It sits at the anchor that names the domain plus 1-2 concrete actions without being comprehensive, and is not generic enough to score 2.	3 / 5
Completeness	Clearly answers both 'what' ('How to delegate all AI work to the agent chat') and 'when' via an explicit 'Use when...' clause with multiple concrete trigger phrases. This matches the 5 anchor exactly and exceeds the 4 anchor where 'when' is present but less explicit.	5 / 5
Trigger Term Quality	Strong natural triggers a developer would actually say — 'delegating AI work from UI or scripts', 'user asks for agent behavior', 'LLM-powered features', 'inline LLM calls' — with good keyword coverage. It falls short of 5 because it lacks synonyms and extension-style variants, and is well above the 3 anchor's thin keyword coverage.	4 / 5
Distinctiveness Conflict Risk	The 'delegate all AI to the local agent chat' niche is mostly distinct with clear triggers, but the description does not disambiguate the local agent from a different agent (the A2A case), creating minor overlap risk with the related a2a-protocol skill. Not a 5 because of that unresolved boundary, clearly above the 3 anchor.	4 / 5
	Total	16 / 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
metadata_version	'metadata.version' is missing	Warning
metadata_field	'metadata' should map string keys to string values	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	13 / 16 Passed

Repository: BuilderIO/agent-native
Path: .agents/skills/delegate-to-agent/SKILL.md
Commit: cc2a915

Reviewed: 5 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.