Name: delegate-to-agent
Rating: 63.2 (1 reviews)
Author: BuilderIO

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

79%

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

Fix and improve this skill with Tessl

tessl review fix ./community-templates/account-expert/.agents/skills/delegate-to-agent/SKILL.md

The canonical home for this skill is delegate-to-agent in BuilderIO/agent-native

Quality

Content

75%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 policy skill: every delegation path has copy-paste code and explicit decision criteria. The main costs are redundancy in the message/context guidance and an inlined sub-agent section that could live in a reference file.

Suggestions

Consolidate the message/context guidance into the table plus one short paragraph; the section, the surface table, and the later form/auto-submit paragraphs repeat the same boundary three times.

Move the Agent Teams sub-agent material (briefing contract, fan-out, synthesis) into a references/sub-agents.md and keep a 5-10 line summary with a link in SKILL.md, trimming roughly 40 lines from the main file.

Add a short note on handling failed or interrupted agent runs (e.g. retry guidance or surfacing the error in the sidebar) to close the workflow-clarity validation gap.

Dimension	Reasoning	Score
Conciseness	Mostly efficient rules and code, but the message/context separation is explained three times (section prose, a four-row surface table, and follow-up paragraphs on forms and auto-submitted turns), and the ~55-line Agent Teams section restates coordination discipline at length, so the body could be meaningfully tightened.	3 / 5
Actionability	Fully executable, copy-paste-ready code for every path (sendToAgentChat foreground and background, agentChat.submit, completeText, useAgentChatGenerating, callAgent) with imports and options, plus decision tables for submit values and context surfaces — the common cases are concretely covered.	5 / 5
Workflow Clarity	Decision branches are unambiguous: the submit-vs-prefill table, the spawn-vs-do-it-yourself criteria, and the four-point sub-agent briefing contract sequence choices clearly. It stops short of 5 because there is no guidance for handling failure states (e.g. a failed or interrupted agent run), though no destructive or batch operations require validation checkpoints.	4 / 5
Progressive Disclosure	Good single-file structure with clear headers, tables, a Related Skills section, and A2A correctly deferred to the a2a-protocol skill. No bundle files exist, and the semi-distinct Agent Teams/sub-agent section (~55 lines) is inlined in SKILL.md rather than split into a reference file, which is the main organization gap.	4 / 5
	Total	16 / 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 what and multiple concrete when-triggers in natural user language. Its weaknesses are that the stated capability is a single action seen from several angles, and the trigger terms do not cover the sub-agent, Agent Teams, and A2A material the body actually contains.

Suggestions

Add trigger terms for the body's secondary topics, e.g. 'when spawning sub-agents or fanning out Agent Teams work' and 'when calling another agent via A2A', so those sections are reachable from the description.

Distinguish this skill from a2a-protocol in the description itself, e.g. 'Use for the local agent chat; use a2a-protocol for remote agents', to reduce overlap risk.

Enumerate one or two more concrete actions (e.g. 'prefill the sidebar composer from app code', 'detect agent completion') to raise capability specificity.

Dimension	Reasoning	Score
Specificity	Names the domain clearly ('delegate all AI work to the agent chat') with a couple of concrete action facets ('delegating AI work from UI or scripts', 'sending messages to the agent from application code'), but every facet is a variation of the same single capability rather than a list of several distinct actions, matching the '1-2 concrete actions, not comprehensive' anchor.	3 / 5
Completeness	Explicitly answers both 'what' ('How to delegate all AI work to the agent chat') and 'when' with multiple concrete trigger clauses ('Use when delegating AI work from UI or scripts... when a user asks for agent behavior or LLM-powered features, when tempted to add inline LLM calls...'), matching the top anchor.	5 / 5
Trigger Term Quality	Includes natural phrases a user would say ('delegating AI work', 'agent chat', 'LLM-powered features', 'inline LLM calls', 'agent behavior') across several scenarios, but misses trigger terms for major parts of the body content such as sub-agents/Agent Teams, A2A, and background runs.	4 / 5
Distinctiveness Conflict Risk	Has a clear niche (delegating to the local agent chat vs inline LLM calls) with distinct triggers, but overlaps with the closely related a2a-protocol skill without distinguishing local-agent vs remote-agent delegation in the description itself.	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: community-templates/account-expert/.agents/skills/delegate-to-agent/SKILL.md
Commit: a941a2e

Reviewed: 2 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.