SKILLEMALL.ai
GitHub ↗ RU Sign in

BC subagent-architecture

Advanced patterns for specialized subagent orchestration with production-ready reference implementations. Security isolation, phased implementation, peer collaboration, and cost-aware spawning.

modbender/skill-library-mcp Agent Skills author: modbender MIT 20 files · 1 script body ≈ 15 985 tokens Open the sourcegithub.com↗ analyzed 2 d ago

Advanced patterns for specialized subagent orchestration with production-ready reference implementations.

As a process C 55/100 · Has gaps — weak spots: result and completion, when it triggers, inputs and preconditions

ReferenceDiscordAI and agentstype and topics are labelled automatically from the skill text
Runs in: OpenClaw Claude / Claude Code ChatGPT / Codex Gemini CLI Cursor GitHub Copilot / VS Code Qwen Code Kimi Code CLI Z.ai (GLM) DeepSeek (Deep Code) Mistral Vibe LangChain (deepagents) Goose OpenCode
JSON
Technical rating
B
87/100
safety, quality, tests
Safety 60%
98
Quality 40%
70
Run on models
none yet
Process rating
C
55/100
Has gaps
Inputs and preconditions w 11
0
When it triggers w 12
20
Running it twice w 4
30
the three weakest of ten parameters · all ten

How to improve

  1. The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
For the model run — optional
  • Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
  • A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.

Guard findings · 2

✓ No critical or high findings

Medium and low: 2
  • low Secrets in code secret-password-literal examples/security-proxy-usage.js:37
    Hard-coded password / key literal (may be an example) (placeholder value)
    api_key: 'sk-a…456',  // Will be redacted
    placeholder
  • low Risky intent intent-offensive-security templates/peer-review-specialist.md:9
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (test fixture / example file)
    - Specialists may have domain expertise you lack (legal review, accessibility audit, security pentesting)
    fixture

Files scanned: 19. Evidence is masked. Grey chips explain why severity was lowered.

Against the Agent Skills spec

  • warning body-long SKILL.md body ≈ 15985 tokens (recommended < 5000); move details to references/
  • note frontmatter-key unknown frontmatter key "difficulty"
  • note frontmatter-key unknown frontmatter key "requirements"
  • note frontmatter-key unknown frontmatter key "credentials_required"
  • note frontmatter-key unknown frontmatter key "external_integrations"
  • note frontmatter-key unknown frontmatter key "package_type"
  • note frontmatter-key unknown frontmatter key "auto_executes"

Process rating: all ten parameters 55/100

  • 0Inputs and preconditions. Does not say what the process needs to start
  • 20When it triggers. No condition that starts the skill
  • 30Running it twice. 13 mutating operations with no state check
  • 40Result and completion. Does not say what the result is
  • 40Execution cost. Instruction body is 15985 tokens: crowds the task out of the window
  • 60Tools and files. Uses tools (bash, web, node) that frontmatter does not declare
  • 85Steps. 407 steps, 2 vague phrases
  • 100Failures and branches. 2 branches, has a failure section
  • 100Consistency. Name and required fields are in place
  • 100Progress reporting. Reports progress
  • medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
  • low 24 top-level sections: this looks like several domains in one skill

Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.

Quality signals

  • +5Description has no quoted example phrases that should trigger the skill
  • +4Description does not say when NOT to use the skill (false activations)
  • +3Output format is not stated: the model decides each time
  • -228 emoji in the instructions: noise for the model
  • +2Single-language instructions
  • +3Description length 193: enough signal without eating the budget
  • +4Structure: 85 headings
  • +3Step-by-step instructions: 407 items
  • +4Has examples (38 code blocks)
  • +1License stated

Quality base 70; lint remarks subtract, signals add up to 100. Result: 70.