do-in-parallel

Run independent tasks concurrently across multiple files or targets using parallel sub-agents, with per-task model selection and LLM-as-a-judge verification. Use when tasks do not depend on each other and can run side by side.

Quality

75%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Critical

Do not install without reviewing

Fix and improve this skill with Tessl

tessl review fix ./plugins/sadd/skills/do-in-parallel/SKILL.md

Quality

Content

66%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 and its workflow is rigorously sequenced with validation and feedback loops, but it is severely over-long for an orchestrator recipe: heavy cross-section repetition and three enormous inlined examples inflate the token budget and bury the core procedure.

Suggestions

Extract the three worked examples (Phases 3.5–5.2 blocks) into reference files (e.g. references/examples-repeatable.md, examples-shared.md, examples-independent.md) and link to them from SKILL.md, leaving only a compact overview inline.

De-duplicate the rules currently repeated in 'RED FLAGS', 'ALWAYS/NEVER', the Process phases, and 'Best Practices' into a single canonical section that the phases link back to.

Trim or collapse the repeated meta-judge/judge prompt templates into one parameterized template referenced by grouping type, instead of re-emitting the full block in every example.

Dimension	Reasoning	Score
Conciseness	At ~2350 lines the body restates the same rules across 'RED FLAGS', 'ALWAYS/NEVER', the Process phases, and 'Best Practices', and repeats near-identical giant prompt templates across three examples — noticeably verbose with substantial padding despite accurate content.	2 / 5
Actionability	Provides fully executable, copy-paste-ready Task tool dispatch blocks, complete meta-judge/judge prompt templates, concrete model-tier tables, and per-phase checklists covering the common grouping cases.	5 / 5
Workflow Clarity	Phases 1–6 are clearly sequenced with an independence validation checklist, judge verdict/score checkpoints, explicit retry/escalation feedback loops, and a decision tree — the batch operation has thorough validation.	5 / 5
Progressive Disclosure	No references/scripts/assets bundle exists; everything — including ~1300 lines of three sprawling examples and repeated full prompt templates that clearly belong in separate reference files — is inlined into one monolithic SKILL.md.	2 / 5
	Total	14 / 20 Passed

Description

83%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 strong: it states concrete capabilities, uses third-person voice, and explicitly pairs a clear 'what' with an explicit 'when' trigger clause. It is just shy of top marks on specificity and trigger-term breadth.

Dimension	Reasoning	Score
Specificity	Lists several concrete mechanisms — 'run tasks concurrently', 'per-task model selection', 'LLM-as-a-judge verification' — but stops short of the fully comprehensive coverage of a 5.	4 / 5
Completeness	Explicitly answers both what ('Run independent tasks concurrently...with per-task model selection and LLM-as-a-judge verification') and when ('Use when tasks do not depend on each other and can run side by side') with a concrete trigger clause.	5 / 5
Trigger Term Quality	Includes natural phrases users would say ('independent tasks', 'concurrently', 'in parallel', 'side by side', 'multiple files or targets'); a few synonyms a user might voice are absent.	4 / 5
Distinctiveness Conflict Risk	The 'parallel sub-agents' + 'per-task model selection' + 'LLM-as-a-judge verification' niche is mostly distinct from generic single-agent skills, with only minor overlap risk against broader sub-agent launchers.	4 / 5
	Total	17 / 20 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
skill_md_line_count	SKILL.md is long (2353 lines); consider splitting into references/ and linking	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: NeoLabHQ/context-engineering-kit
Path: plugins/sadd/skills/do-in-parallel/SKILL.md
Commit: 8539779

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.