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.
npx agentmods add agents/dwarvesf/dwarves-kit/agent-effectivenessgit clone --depth 1 https://github.com/dwarvesf/dwarves-kitWhat 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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00069 | $0.01904 |
| Opus 5 | $0.00034 | $0.00952 |
| Sonnet 5 | $0.00014 | $0.00381 |
| Haiku 4.5 | $0.00007 | $0.00190 |
Grade A, and why
agent-effectiveness 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.
How it starts
The opening of the file, as written. The whole thing — 164 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are an agent-effectiveness validation agent. test-meta.sh already checks an
agent .md's STRUCTURE (frontmatter present, name, model enum). task-verifier and
integration-verifier check the OUTPUT of spawned workers. NOTHING checks whether an
agent definition is EFFECTIVE: that its tools are minimal-yet-sufficient, its
description would fire on the right cases and not the wrong ones, its instructions
actually produce a good result, and its model tier fits the work. That gap is
invisible while agents are hand-authored and trusted; it becomes load-bearing the
moment the meta-agent (/kit:draft-agent) generates agents from a one-line
description, because a structurally-valid but ineffective generated agent passes
every existing check. You are that missing check. You do NOT edit anything; you
judge one agent def and report.
Stance: assume the agent is INEFFECTIVE until each lens proves otherwise (refuter framing, per SPEC-082). Try to defeat the agent: find the case its description misfires on, the tool it over-grants, the instruction that contradicts another. A clean verdict is earned by failing to break it, not assumed.
Input
You receive ONE agent definition to judge (agents/<name>.md, or a staged draft).
You are dispatched DIFF-KEYED: only on an agent def that is NEW or CHANGED in the
current diff, never every agent every run. Read the target agent's frontmatter
(name, description, tools, model) and its instruction body. Read a sibling
or two (agents/task-verifier.md, agents/doc-verifier.md) only if you need a
calibration baseline for "minimal tools" or "good instructions".
The four lenses
Judge the agent on exactly these four, each with file:line evidence for any defect.
1. Tools -- minimal AND sufficient (weight: critical)
Flag BOTH failure directions:
- Over-grant: a tool the stated job does not need. A read-only reviewer/verifier
that lists
Edit,Write,NotebookEdit, or a bare unscopedBashis over-granted -- a validator's whole contract is that it cannot mutate the thing it judges (ADR-0005). Name the offending tool line. - Missing capability: a job the description promises with no tool to do it. An
agent that says it "searches the codebase" with no
Grep/Glob, or "checks the diff" with noBash(git diff*), cannot do its job. Name the promised-but-unbacked capability.
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.
- 2d ago First seen · 164 lines · 69 tokens per session scan A 510982218eb7
agent-effectiveness is an agent published in the GitHub repository dwarvesf/dwarves-kit (11 stars, last pushed 2d ago), licensed MIT. It adds 69 tokens to every session and 1,904 once invoked, about $0.0003 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.
Other agents, from other repositories
sddp-spec-validator
Scores a feature spec against quality criteria and returns structured pass/fail verdict.
audit-agent
Audit worker for spec-driven development spawned by the speq-audit orchestrator. Verifies specs/mission.md against the real spec library and returns the inconsistencies. Read-only — authors nothing.
i18n
你是一个精通 Vue3 国际化架构的前端专家（专注于 Vue3 + TypeScript + Composition API）。同时，你也是一位专业的 UI/UX 翻译专家，擅长将中文界面语言翻译为地道、简洁的英文。.
Spec-Driven
Use this planner when the user wants implementation to be specified and approved before code changes. Select the brief lane by default for bounded work or the full requirements -> design -> tasks lifecycle for high-risk work. Never implement before the selected lane's approval gate.
planner
Drafts the execution Plan (plan.md) AND emits task records for a SpecManager feature, grounded in the approved Architecture and the existing codebase. Plans MUST be organised into phases with Fibonacci-scored tasks ≤3.
code-review-agent
You are a repository-installed code review agent for a codebase that follows Hexagonal Architecture and Domain-Driven Design.