dispatching-parallel-agents
Use when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies
Install with Tessl CLI
npx tessl i github:projectbluefin/dakota --skill dispatching-parallel-agentsOverall
score
59%
Does it follow best practices?
If you maintain this skill, you can automatically optimize it using the tessl CLI to improve its score:
npx tessl skill review --optimize ./path/to/skillAgent success when using this skill
Validation for skill structure
Evaluation results
100%
34%Multi-Service Incident Triage
Independent domain identification and parallel dispatch decision
Independent domain recognition
66%
100%
Parallel dispatch recommended
80%
100%
One investigator per domain
100%
100%
Auth service excluded from parallel work
100%
100%
Independence reasoning
20%
100%
No shared state concern noted
20%
100%
Scope per investigator defined
90%
100%
Integration step included
50%
100%
Without context: $0.2043 · 1m 23s · 7 turns · 8 in / 3,559 out tokens
With context: $0.2795 · 1m 33s · 13 turns · 11 in / 4,401 out tokens
100%
10%Automated Test Failure Investigation
Quality of parallel agent prompt construction
One prompt per test file
100%
100%
Specific test names included
100%
100%
Error messages or expected/received included
100%
100%
Scope constraint present
58%
100%
Clear output specified
58%
100%
No broad 'fix everything' prompts
100%
100%
Prompts are self-contained
100%
100%
No timeout-only fix instruction
100%
100%
Prompts are distinct
100%
100%
Without context: $0.1540 · 46s · 6 turns · 55 in / 2,533 out tokens
With context: $0.3108 · 1m 19s · 13 turns · 261 in / 3,982 out tokens
100%
Parallel Agent Results Integration
Post-dispatch integration review and conflict detection
Agent summaries reviewed
100%
100%
Conflict identified
100%
100%
Conflict explanation
100%
100%
StorageAdapter independence noted
100%
100%
Integration order recommended
100%
100%
Spot check item identified
100%
100%
Full suite verification recommended
100%
100%
No timeout-increase recommendation
100%
100%
Without context: $0.1997 · 1m 4s · 8 turns · 57 in / 3,190 out tokens
With context: $0.3578 · 1m 38s · 16 turns · 14 in / 4,759 out tokens
- Evaluated
- Agent
- Claude Code
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.