CtrlK
BlogDocsLog inGet started
Tessl Logo

00-meta-chain-flow-150

[00] META. Orchestrate skills into dynamic chains for complex tasks. Analyzes the task, discovers available skills, builds an optimal chain, explains why each skill is needed, and executes step-by-step with user confirmation. Use for any complex task requiring multiple thinking/research/analysis steps. Triggers on "plan this", "how to approach", "what's the strategy", "build a plan", or any multi-step problem requiring skill orchestration.

89

1.56x

Quality

88%

Does it follow best practices?

Impact

89%

1.56x

Average score across 3 eval scenarios

SecuritybySnyk

Passed

No known issues

SKILL.md
Quality
Evals
Security

Evaluation results

86%

38%

Real-Time Collaboration Feature Planning

Skill discovery and chain output format

Criteria
Without context
With context

Scans codex skills dir

100%

100%

Scans legacy skills dirs

50%

100%

Scans home skills dir

0%

100%

Chain-Flow header present

0%

0%

Task Analysis section

28%

100%

Skills Discovered section

75%

75%

Proposed Chain section

100%

100%

WHY justifications

100%

100%

Execution Plan section

50%

75%

Confirmation policy noted

0%

100%

Adaptation policy noted

0%

100%

New Feature pattern followed

50%

100%

Without context: $0.3438 · 2m 2s · 17 turns · 21 in / 5,943 out tokens

With context: $0.4422 · 2m 12s · 18 turns · 321 in / 6,640 out tokens

86%

38%

Production Database Timeout Investigation

Chain execution format and adaptation

Criteria
Without context
With context

Step X/N header format

30%

100%

Current Skill label

0%

75%

Purpose label per step

25%

100%

Step complete marker

0%

100%

Continuation prompt per step

11%

44%

Bug Investigation pattern

88%

100%

Impact mapping step

71%

100%

Deep analysis step

85%

100%

Chain adaptation section

90%

100%

Adaptation shows specific change

88%

100%

Confirmation after adaptation

0%

25%

150% rule applied

75%

87%

Without context: $0.2616 · 1m 42s · 12 turns · 19 in / 4,898 out tokens

With context: $0.5496 · 3m 26s · 18 turns · 53 in / 10,197 out tokens

96%

19%

Analytics Platform Database Architecture Decision

Chain patterns and failure recovery

Criteria
Without context
With context

Architecture Decision pattern

90%

100%

WHY for each skill

100%

100%

Execution trace format

75%

100%

Current skill and purpose

100%

100%

Chain Adaptation section

88%

100%

Adaptation specifies chain change

100%

100%

Adaptation confirmation request

0%

85%

Failure mode 1 with recovery

100%

100%

Failure mode 2 with recovery

100%

100%

150% core element

62%

100%

150% enhancement element

87%

100%

Chain flexibility noted

0%

62%

Without context: $0.2803 · 1m 59s · 12 turns · 19 in / 5,376 out tokens

With context: $0.3121 · 2m 12s · 11 turns · 274 in / 5,936 out tokens

Repository
majiayu000/claude-skill-registry-data
Evaluated
Agent
Claude Code
Model
Claude Sonnet 4.6

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.