skill-parallel-agents

Decompose large tasks across parallel agents — use for migrations, multi-file refactors, or batch work

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 ./.claude/skills/skill-parallel-agents/SKILL.md

Quality

Content

56%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 highly actionable — executable orchestrate.sh commands, explicit phase sequencing, provider routing rules, and validation gates — but it is heavily padded with compliance rhetoric, duplicated guidance, ASCII art, and marketing tone. It is also a monolithic 640-line document that should split its command reference and provider configuration into separate reference files.

Suggestions

Cut the "MANDATORY COMPLIANCE" block, ASCII-art diagrams, and the "When NOT to Use" section (which duplicates the frontmatter trigger), and merge the two cost tables to roughly halve the body's token footprint.

Move the Command Reference, Provider-Aware Routing/configuration, and OpenRouter/OrcaRouter fallback sections into reference files (e.g. references/commands.md, references/providers.md) linked from a concise overview, enabling progressive disclosure.

State explicitly what Claude should do with each phase's output (probe-synthesis-*.md, grasp-consensus-*.md) beyond passing --context, and standardize script paths on ${CLAUDE_PLUGIN_ROOT} with the ${HOME} fallback used everywhere.

Dimension	Reasoning	Score
Conciseness	The 640-line body is noticeably verbose and repetitive: a "MANDATORY COMPLIANCE" block, ASCII-art diamond and grapple diagrams, a "When NOT to Use" section duplicating the frontmatter trigger list, provider cost tables shown twice, and marketing prose. It stays above a 1 because it is operational reference rather than explanation of concepts Claude already knows.	2 / 5
Actionability	Commands are concrete and copy-paste ready throughout, with realistic prompts, option tables, a full provider-config YAML, and sample detect-providers output. Falls short of 5 because expected outputs of the main workflow commands are not shown and script paths are inconsistent (hardcoded ${HOME}/.claude-octopus in most examples vs ${CLAUDE_PLUGIN_ROOT} in one).	4 / 5
Workflow Clarity	The Double Diamond sequence is explicit with --context chaining between phases, and validation checkpoints are present (prerequisites provider check with routing rules, preflight, and the tangle quality-gate table with remediation guidance in Troubleshooting). Not a 5 because what Claude should do with each phase's synthesis output is largely implicit in script behavior rather than stated.	4 / 5
Progressive Disclosure	No bundle files exist, and everything — the full command reference, provider-tier tables, configuration examples, and OpenRouter/OrcaRouter fallback docs — is inlined in one 640-line SKILL.md. Section headers are abundant and navigation works (so not a 2), but substantial content that belongs in separate reference files is inlined (so not a 4).	3 / 5
	Total	13 / 20 Passed

Description

75%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 concise and cleanly answers both what the skill does and when to use it, with reasonably natural trigger terms. Its main weakness is distinctiveness: the wording suggests generic agent parallelization rather than the multi-provider orchestration the skill actually performs, creating overlap risk with built-in delegation features.

Dimension	Reasoning	Score
Specificity	"Decompose large tasks across parallel agents" names the domain and one concrete action, with use cases ("migrations, multi-file refactors, or batch work") listed, but action coverage is not comprehensive — only a single decompose action is described.	3 / 5
Completeness	Explicitly answers both: what ("Decompose large tasks across parallel agents") and when ("use for migrations, multi-file refactors, or batch work") with concrete trigger phrases. The when-clause is explicit and specific, so not a 4.	5 / 5
Trigger Term Quality	"migrations", "multi-file refactors", and "batch work" are natural phrases users would say when needing this skill. Missing common synonyms and variations such as "run in parallel", "parallelize", or "fan out", so not a 5.	4 / 5
Distinctiveness Conflict Risk	"parallel agents" and "large tasks" overlap broadly with built-in subagent/Task-tool delegation and generic orchestration skills, and the description omits the skill's actual niche (multi-provider external CLIs such as Codex and Antigravity). The migration/refactor triggers give some specificity, so not lower.	3 / 5
	Total	15 / 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 (675 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: nyldn/claude-octopus
Path: .claude/skills/skill-parallel-agents/SKILL.md
Commit: 0093dab

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.