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 skills/typedef-ai/ade-bench-plugin/setupnpx skills add typedef-ai/ade-bench-plugin --skill setupgit clone --depth 1 https://github.com/typedef-ai/ade-bench-pluginWhat 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 | $0.00079 | $0.02544 |
| Opus 5 | $0.00039 | $0.01272 |
| Sonnet 5 | $0.00016 | $0.00509 |
| Haiku 4.5 | $0.00008 | $0.00254 |
Grade C, and why
setup scanned grade C with 2 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.
Downloads and executes remote codehighSupply chain
curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.
- **uv**: `curl -LsSf https://astral.sh/uv/install.sh | sh` (then restart the shell or `source ~/.zshrc`). Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- **uv**: `curl -LsSf https://astral.sh/uv/install.sh | sh` (then restart the shell or `source ~/.zshrc`). How it starts
The opening of the file, as written. The whole thing — 227 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ADE-Bench Setup
The plugin generates task definitions. ade-bench (a separate Python project) is what actually executes them inside Docker against the user's data. This skill makes ade-bench available so generated tasks can be run.
Default install location: ~/.ade-bench. The skill git clones ade-bench there, runs uv sync to install Python deps, and uv tool install . to put the ade CLI on PATH.
Step 1: Parse Arguments
--path <install-path>: Where to install ade-bench. Default:~/.ade-bench.
Resolve ~ to the user's home directory. Use the resolved absolute path in every step below.
Step 2: Check Prerequisites
Run these in parallel and collect the results:
- git:
command -v git - uv:
command -v uv - docker:
command -v docker
If any of the three is missing, stop and tell the user what's missing and how to install it:
- git: Use the system package manager (
brew install git,apt install git, etc.). - uv:
curl -LsSf https://astral.sh/uv/install.sh | sh(then restart the shell orsource ~/.zshrc). - docker: Recommend Docker Desktop, Colima, or OrbStack. Don't try to install or start Docker. The user must do this.
If docker is installed but not running, that's fine for setup — flag it but proceed. Docker only needs to be running when tasks actually execute.
Step 3: Detect Existing Installation
Two checks, in order:
- Already installed via this skill?
test -f <install-path>/pyproject.toml && grep -q 'name = "ade-bench"' <install-path>/pyproject.toml - Already on PATH from somewhere else?
command -v ade
If either succeeds:
- Run
ade --helpto confirm the CLI works. - Tell the user where ade-bench was found and that no install is needed.
- Skip to Step 7 (DuckDB database download — still verify it's been done) and continue from there.
If <install-path> exists but is not a valid ade-bench checkout (different repo, partial clone, leftover files), stop and ask the user whether to remove and reinstall or pick a different path. Do not silently overwrite.
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 · 227 lines · 79 tokens per session scan C 4ee2ebfbad05
setup is a skill published in the GitHub repository typedef-ai/ade-bench-plugin (3 stars, last pushed 3mo ago), licensed MIT. It adds 79 tokens to every session and 2,544 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it C with 2 findings (downloads and executes remote code, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
pr-verify
Verify a Docglow change actually works before submitting or merging a PR. Runs the conformance suite, then a behavioral verification pass (flag matrix, artifact-join spot checks, pipeline contract sweep, payload budget). Use when reviewing a PR, self-reviewing a branch before opening a PR, or when asked to "verify…
dbt-agent-readiness
Audit a dbt project for agent-readiness: what would an AI agent get wrong if you pointed it at this data today? Produces a prioritized report organized by failure modes (wrong numbers, wrong table, wrong column, can't join, query fails). Scales via two-pass architecture with parallel subagents. Each subagent reads its…
authoring-meta-context
Authors and extracts dbt MetricFlow meta context — structured YAML meta: blocks that encode business knowledge (thresholds, investigation paths, SLAs, relationships) alongside metric definitions so AI agents answer analytical questions accurately. Use when user asks about context cards, meta context blocks, dbt metric…
animation-best-practices
CSS and UI animation patterns for responsive, polished interfaces. Use when implementing hover effects, tooltips, button feedback, transitions, or fixing animation issues like flicker and shakiness.
dbt-doctor
Static analysis and health checks for dbt projects. Use before committing SQL/YAML or when enforcing CI quality gates.
dbt-expert
Expert-level dbt (data build tool), models, tests, documentation, incremental models, macros, and Jinja templating.