CtrlK
BlogDocsLog inGet started
Tessl Logo

dispatching-parallel-agents

Use when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies

53

Quality

58%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No known issues

Fix and improve this skill with Tessl

tessl review fix ./plugins/superpowers/skills/dispatching-parallel-agents/SKILL.md

The canonical home for this skill is dispatching-parallel-agents in obra/superpowers

SKILL.md
Quality
Evals
Security

Quality

Content

85%

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 with concrete dispatch calls, a reusable agent-prompt template, and an explicit verification workflow. Its main weakness is conciseness: it re-explains obvious rationale and repeats the same 'when not to use' and integration guidance across multiple sections.

Suggestions

Collapse the duplicate 'Don't use when' and 'When NOT to Use' sections into one, and merge the overlapping integration guidance from 'Review and Integrate', 'Verification', and 'Real-World Impact' into a single verification section.

Trim explanatory rationale such as 'investigating them sequentially wastes time' that restates why parallelism helps, since Claude already understands this.

Consider moving the detailed 'Real Example from Session' (with specific 2025-10-03 dates and per-agent results) into a referenced example file to keep the overview lean, though this is optional given no bundle files currently exist.

DimensionReasoningScore

Conciseness

The body is mostly efficient with concrete examples, but it explains concepts Claude already knows (why sequential investigation 'wastes time') and duplicates material across 'Don't use when', a later 'When NOT to Use' section, and the repeated integration guidance in 'Review and Integrate', 'Verification', and 'Real-World Impact', fitting 'mostly efficient but could be tightened' rather than the lean level 3.

2 / 3

Actionability

It provides a fully usable agent-prompt template ('Fix the 3 failing tests in src/agents/agent-tool-abort.test.ts: ... Return: Summary of what you found and what you fixed.'), concrete Task(...) dispatch calls, and specific correct/incorrect mistake patterns, matching 'fully executable; specific examples; copy-paste ready' rather than the pseudocode level 2.

3 / 3

Workflow Clarity

The four-step 'The Pattern' is clearly sequenced and paired with an explicit verification section ('Run full test suite', 'Check for conflicts', 'Spot check - Agents can make systematic errors'), giving clear validation checkpoints and feedback for a batch operation, matching the level-3 anchor; it is not level 2 because checkpoints are explicit rather than implicit.

3 / 3

Progressive Disclosure

With no bundle files present, the single self-contained SKILL.md is organized into clearly headed sections (Overview, When to Use, The Pattern, Common Mistakes, Real Example, Verification) with no nested or broken references, so per the rubric's simple-skill note it qualifies for level 3 on well-organized sections; the mild content redundancy is a conciseness concern rather than a navigation defect.

3 / 3

Total

11

/

12

Passed

Description

32%

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 essentially a trigger clause with no statement of what the skill actually does, leaving completeness as the weakest dimension. Its terms are concrete but technical and omit the most natural parallelism phrasings users would say.

Suggestions

Add a leading capability clause stating what the skill does, e.g. 'Dispatches parallel sub-agents to work on independent tasks concurrently. Use when facing 2+ independent tasks...'

Include natural user phrasings such as 'in parallel', 'concurrently', or 'at the same time' alongside the technical 'without shared state or sequential dependencies' phrasing.

Tighten distinctiveness by naming the agent-dispatch mechanism (e.g. 'one sub-agent per independent problem domain') so it cannot be confused with general task-management skills.

DimensionReasoningScore

Specificity

The description names a concrete domain ("2+ independent tasks", "without shared state or sequential dependencies") but lists no actual capabilities or actions, matching the 'names domain and some actions, but not comprehensive' anchor rather than the multi-action level 3.

2 / 3

Completeness

It answers only the 'when' ("Use when facing 2+ independent tasks...") and never states what the skill does, so the 'what' is missing, which matches the 'missing what OR when' level-1 anchor; it is not level 2 because level 2 requires the 'what' to be present.

1 / 3

Trigger Term Quality

It opens with a natural "Use when" trigger, but the terms ('independent tasks', 'shared state', 'sequential dependencies') are technical and miss common natural phrasings a user would say like 'in parallel', 'concurrently', or 'at the same time', so it is 'some relevant keywords but missing common variations' rather than full coverage.

2 / 3

Distinctiveness Conflict Risk

The parallel/independence niche is fairly distinct from file-format skills, yet the generic phrase 'independent tasks' could overlap with general task-management skills, fitting 'somewhat specific but could still overlap' rather than the conflict-proof level 3.

2 / 3

Total

7

/

12

Passed

Validation

100%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
openai/plugins
Reviewed

Table of Contents

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.