parallel-agents

Multi-agent orchestration patterns. Use when multiple independent tasks can run with different domain expertise or when comprehensive analysis requires multiple perspectives.

Quality

65%

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/parallel-agents/SKILL.md

Quality

Content

55%

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

The content is clearly organized and reasonably lean, with sequenced orchestration patterns and a synthesis template that give usable structure. Its weaknesses are non-executable (natural-language) instructions, missing validation/verification checkpoints for multi-agent runs, and reference material that sits inline instead of being split into bundled files.

Suggestions

Replace natural-language prompt blocks with concrete, executable invocation syntax or exact tool-call patterns so the guidance is copy-paste ready.

Add explicit validation/verification checkpoints (e.g., 'confirm each agent returned findings before synthesis; re-invoke on empty output') to satisfy the feedback-loop requirement for multi-step coordination.

Move the large agent catalog and synthesis template into bundled reference files (e.g., references/AGENTS.md, references/SYNTHESIS.md) and link to them one level deep, leaving SKILL.md a lean overview.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient and well-organized into labeled sections, but the large 'Available Agents' table (17 custom agents plus a built-in table) and verbose pattern/synthesis scaffolding include content that could be tightened; it doesn't pad with concepts Claude already knows, yet some repetition (good-for/not-for vs when-to-use) is unnecessary.	2 / 3
Actionability	It provides concrete-ish guidance via named agents and sequenced patterns, but the 'code blocks' are natural-language prompts and template placeholders ('Use the security-auditor agent...', 'Resume agent [agentId]') rather than executable, copy-paste-ready instructions or commands, matching the level-2 'pseudocode/incomplete' anchor.	2 / 3
Workflow Clarity	Multi-step patterns are clearly sequenced (Discovery -> Analysis -> Synthesis) with a synthesis template, but there are no explicit validation checkpoints or error-recovery feedback loops; for an orchestration skill whose whole point is coordinating fragile multi-agent runs, the absence of verification steps caps this at 2 per the rubric's feedback-loop guidance.	2 / 3
Progressive Disclosure	The body is well-sectioned with no bundle files present (references/scripts/assets do not exist), so there is no nested-reference problem; however, large reference-like material (the 17-agent table, synthesis template) is inline rather than split into separate files, leaving it somewhat dense rather than a lean overview with one-level-deep pointers.	2 / 3
	Total	8 / 12 Passed

Description

75%

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 clear and complete on the what/when axis with an explicit trigger clause, but it stays at the level of situations rather than listing concrete operations, and its triggers are somewhat generic, leaving modest conflict risk. It is solid but not exemplary on specificity and distinctiveness.

Suggestions

Add concrete verbs/actions (e.g., 'Spawn, coordinate, and synthesize outputs from parallel specialist agents') to lift specificity from situations to enumerated operations.

Sharpen trigger terms with user-natural phrasings and their variations (e.g., 'fan out agents', 'parallel review', 'multi-agent analysis') to improve coverage and reduce overlap.

Tighten distinctiveness by naming the differentiator from single-agent work (e.g., 'when parallel execution of independent expert agents beats one agent') so it won't trigger for ordinary analysis.

Dimension	Reasoning	Score
Specificity	The description names a domain ('Multi-agent orchestration patterns') and some concrete situations ('multiple independent tasks can run with different domain expertise', 'comprehensive analysis requires multiple perspectives') but does not enumerate specific concrete actions the way the level-3 anchor ('Extract text and tables... fill forms, merge documents') does; it describes when rather than discrete operations.	2 / 3
Completeness	It explicitly answers both 'what' ('Multi-agent orchestration patterns') and 'when' via an explicit 'Use when...' clause ('Use when multiple independent tasks can run... or when comprehensive analysis requires multiple perspectives'), matching the level-3 anchor requiring both with an explicit trigger.	3 / 3
Trigger Term Quality	It offers some natural trigger phrases ('multiple independent tasks', 'comprehensive analysis', 'multiple perspectives') that a user might voice, but coverage is thin and lacks common variations a user would naturally say; it does not reach the broad keyword coverage of the level-3 example.	2 / 3
Distinctiveness Conflict Risk	'Multi-agent orchestration' is a recognizable niche, but 'comprehensive analysis' and 'multiple perspectives' are broad enough to overlap with general review or analysis skills, so it could conflict with similar skills rather than having clearly distinct triggers.	2 / 3
	Total	9 / 12 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: vudovn/ag-kit
Path: .agents/skills/parallel-agents/SKILL.md
Commit: 04514f0

Reviewed: about 11 hours 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.