delegating-work

Which tier — main thread vs. a cheaper subagent — owns a piece of work, decided at the moment you're about to start it. Use before writing an implementation yourself, driving a browser, babysitting a PR, running a test/fix loop, or doing a mechanical multi-file sweep on the main thread.

Quality

75%

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 ./.agents/skills/delegating-work/SKILL.md

Quality

Content

67%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 a well-structured, actionable decision guide with a clear table and concrete tier defaults. It is somewhat over-padded by redundant italicized user quotes and a time-sensitive date that could be trimmed without losing clarity.

Suggestions

Trim the duplicated italicized user quotes (keep one or two as evidence, drop the rest) and remove the date stamp to improve token efficiency.

Add a concrete one-line example of how to split a mechanical sweep into disjoint file sets per subagent.

Briefly inline the handoff boundary to efficient-frontier (e.g. the briefing contract fields) so the end-to-end flow is complete within the file.

Dimension	Reasoning	Score
Conciseness	Mostly efficient and assumes Claude's competence (no basic-concept explanations), but each bullet restates its rule in an italicized user quote and includes a time-sensitive date ('most recently today (2026-07-31)'), which is trimmable padding.	3 / 5
Actionability	A concrete decision table maps work-kinds to tiers, with specific defaults ('Haiku for bulk/mechanical work, Sonnet for real coding judgment') and explicit spawn instructions ('Spawn a subagent to run Playwright / Chrome DevTools MCP and report back'); minor gaps in the parallel-edit mechanics.	4 / 5
Workflow Clarity	Clear decision flow (guard → table → delegate) with a re-read-before-edit checkpoint in the parallel-edits section, but the actual delegation workflow is handed off to related skills, leaving a minor end-to-end gap within this file.	4 / 5
Progressive Disclosure	Well-organized sections in a self-contained file with a clean one-level 'Related skills' pointer; no bundle files exist so there are no nested references, though the body exceeds the 50-line simple-skill threshold.	4 / 5
	Total	15 / 20 Passed

Description

76%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 concrete, complete, and clearly scoped to a distinct niche, with explicit what/when trigger guidance. Its main weaknesses are second-person voice (a specificity penalty) and trigger terms that read as internal workflow jargon rather than natural user utterances.

Suggestions

Rewrite in third person ('Decides which tier owns a piece of work...') to avoid the second-person specificity penalty.

Add natural user-facing trigger phrasings (e.g. 'parallelize', 'delegate this', 'use a cheaper model') alongside the internal jargon.

Sharpen the distinction from the related delegate-to-agent / efficient-frontier skills to reduce overlap ambiguity at trigger time.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete trigger actions — 'writing an implementation yourself, driving a browser, babysitting a PR, running a test/fix loop, or doing a mechanical multi-file sweep' — which is comprehensive coverage; base 5 reduced to 4 because the description uses second-person voice ('you're about to start it', 'yourself') rather than third person.	4 / 5
Completeness	Explicitly answers both 'what' ('Which tier — main thread vs. a cheaper subagent — owns a piece of work') and 'when' ('Use before writing an implementation yourself, driving a browser, babysitting a PR...') with concrete trigger phrases.	5 / 5
Trigger Term Quality	Relevant keywords are present but lean toward internal agent jargon ('main thread', 'subagent', 'tier', 'babysitting a PR') rather than phrases a user would naturally say, and it omits common user utterances like 'parallelize this' or 'delegate this'.	3 / 5
Distinctiveness Conflict Risk	Has a clear niche (tier-ownership decision at the moment of starting work) but minor overlap risk with closely related skills it itself names (delegate-to-agent, efficient-frontier).	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/delegating-work/SKILL.md
Commit: a941a2e

Reviewed: 9 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.