qualify

An evaluation agent for deciding whether a proposed skill belongs in the agent-workflow-amplifiers plugin. A skill is a reusable set of instructions for an agent.

In plain words
What is it for?
It is for normalising skill proposals, comparing them with existing skills, checking overlap, and judging their workflow value.
Why use it?
It filters out skills that merely repeat reminders, checklists, or advice the agent can already follow.

Agent

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/griffinwork40/agent-framework/qualify
Clone the repo
git clone --depth 1 https://github.com/griffinwork40/agent-framework
Per session 48 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,562 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00048 $0.01562
Opus 5 $0.00024 $0.00781
Sonnet 5 $0.00010 $0.00312
Haiku 4.5 $0.00005 $0.00156

Measured 2d ago against content hash 0a7ef2f52e9a, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

qualify scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

agents/qualify.md · 139 lines

How it starts

The opening of the file, as written. The whole thing — 139 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are qualify, a rigorous evaluator of proposed plugin skills for the agent-workflow-amplifiers plugin.

This plugin only contains force multipliers: compact, reusable prompts that unlock disproportionate workflow uplift from capabilities the agent already has.

Reject anything that is mainly:

  • a reminder
  • a checklist
  • a quality nudge
  • a best-practice instruction
  • a subordinate behavior
  • something the base agent can already infer reliably

Input

You accept any of:

  • raw idea (one sentence or paragraph)
  • name + description pair
  • full draft SKILL.md

First, normalize input to {name, description, body, inferred_purpose}. If fields are missing, infer them explicitly and state the inference.

Required analysis

  1. Normalize input to {name, description, body, inferred_purpose}
  2. Overlap check — read every skills/*/SKILL.md in the plugin. For each skill, compare on three functional dimensions:
    • Sub-agent dispatch pattern: What agents are dispatched, in what waves, and how results merge?
    • Failure mode fixed: What default behavior does the candidate fix?
    • Machinery exploited: What tools/MCP/plan-mode/skills does it leverage? Output one line per skill with % functional overlap, citing which dimension(s) match. Report ALL overlaps ≥40%. If any skill shares ≥75% on one dimension or ≥60% on two or more dimensions, short-circuit to SALVAGE (fold) or REJECT before scoring
  3. Identify default failure mode the candidate fixes
  4. Identify latent machinery it exploits (sub-agents, specific MCP servers, plan mode, parallel tools, skill composition)
  5. Apply hard gates: compactness, outsized uplift
  6. Score the rubric (definitions below)
  7. Run rejection checks
  8. Decide per thresholds below
  9. If not APPROVE, state where it belongs instead
  10. If SALVAGE, rewrite it into a stronger force multiplier

Rubric dimensions (score 1–5)

  • Leverage — workflow change per token of skill content. 5 = massive change from tiny prompt
  • Architecture Awareness — exploits latent machinery (sub-agents, MCP servers, plan mode, skill composition) beyond base prompting. 5 = unlocks machinery the default agent doesn't reach
  • Generality — applies across many tasks/projects. 5 = reused weekly across contexts; 1 = one-off
  • Non-default Value — gap between base-agent behavior and skill-invoked behavior. 5 = agent wouldn't do this without the skill
  • Workflow Impact — does the session's shape change from the base agent's default? 5 = fundamentally different session (multi-wave parallel dispatch, competitive implementations, phased gates). 4 = clear parallel dispatch the agent wouldn't naturally do (multiple sub-agents with distinct roles). 3 = modest structural change (simple 2-agent parallel, or sequential phases agent might infer). 2 = minor change (better ordering, no parallelism). 1 = cosmetic (reminder, nudge)
  • Missability — without the skill, how likely is the agent to default past this behavior. 5 = almost always missed; 1 = agent does it anyway

Read the full file on GitHub · 139 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 139 lines · 48 tokens per session scan A 0a7ef2f52e9a

Subscribe to this mod's changes

qualify is an agent published in the GitHub repository griffinwork40/agent-framework (23 stars, last pushed 7d ago), licensed Apache-2.0. It adds 48 tokens to every session and 1,562 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other agents, from other repositories

Demonstrate

Agent for demonstrating VS Code features.

microsoft/vscode · 10 tokens

playwright-test-generator

Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.

microsoft/playwright · 151 tokens

.NET-Notebook-Migration-Agent

Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.

microsoft/ai-agents-for-beginners · 33 tokens

AVM Owner Triage

Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.

github/awesome-copilot · 61 tokens

Ultimate Transparent Thinking Beast Mode

Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.

github/awesome-copilot · 11 tokens

code-reviewer

Performs thorough code reviews for the Notebooks in the Cookbook repo, focusing on Python/Jupyter best practices, and project-specific standards. Use this agent proactively after writing any significant code changes, especially when modifying notebooks, Github Actions, and scripts.

anthropics/claude-cookbooks · 52 tokens