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 skills add amadeus-dlc/amadeus --skill amadeus-grillinggit clone --depth 1 https://github.com/amadeus-dlc/amadeusWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/amadeus-dlc/amadeus/amadeus-grilling)<a href="https://agentmods.dev/skills/amadeus-dlc/amadeus/amadeus-grilling"><img src="https://agentmods.dev/badge/skills/amadeus-dlc/amadeus/amadeus-grilling/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/amadeus-dlc/amadeus/amadeus-grilling"><img src="https://agentmods.dev/badge/skills/amadeus-dlc/amadeus/amadeus-grilling.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00112 | $0.00915 |
| Opus 5 | $0.00056 | $0.00458 |
| Sonnet 5 | $0.00022 | $0.00183 |
| Haiku 4.5 | $0.00011 | $0.00092 |
Grade A, and why
amadeus-grilling 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 6d 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 — 75 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Amadeus Grilling
Purpose
Interview the user relentlessly about every aspect of a plan, design, or decision space until agent and user reach a shared understanding — the design tree worked round by round, every question carrying a recommended answer and its rationale.
The discipline is defined ONCE, in
{{HARNESS_DIR}}/amadeus-common/protocols/grilling-protocol.md (§1 the
upstream skeleton — design tree, rounds, frontier; §2 the Amadeus overlay —
pruning threshold, deferred nodes, circuit breaker; §3 Question Spec
Templates). Read that file first and follow it exactly. This skill adds only
the standalone-specific rules below — it does not re-define the discipline.
Classification
Read-only. This skill never advances the workflow stage pointer and never
emits an audit event. It is safe to run at any point, with or without an
active workflow. The workflow-only obligations in the protocol (questions
file, decision/answer audit logging) do NOT apply here — everything
happens in the conversation.
Standalone rules
- Subject comes from the argument: a file path or a topic. If a file path, read it (and whatever it references) as the starting material for the Investigate step. If no argument was given, ask "What should I grill you about?" as the very first question.
- Discipline is the protocol's §1 and §2 in full: the design tree worked in rounds, the whole pruned frontier asked per round with recommended answers and rationale, facts self-researched with only decisions asked, estimates confirmed with a confidence level, and termination when the pruned frontier is empty or the user says "done". Close with an explicitly confirmed agreement summary that lists the deferred nodes and every unresolved material point.
- Level is the second argument and sets the pruning threshold only
(protocol §2.2) — never a question budget.
Minimal/Standard/Comprehensiveprune progressively less and each arms the circuit breaker at 12 / 24 / 36 rendered questions (protocol §2.4).Freeprunes nothing and has no breaker: every branch of the tree is visited, and the safety valves are the per-round gate and "done". Default to Free when the user names no level — single use matches upstream behaviour. Under Free, say so explicitly in the summary's deferred section ("none — Free prunes nothing") rather than omitting it. - Output is terminal-only. Print the agreement summary (the full
question → decision table, the deferred nodes, and the unresolved
material points) to the terminal.
Write it to a file ONLY when
the user explicitly asks for it to be saved to a path (e.g. "save the
summary to docs/plan-review.md") — the same explicit-request exception
/amadeus-outcomes-packuses. Never write anywhere unprompted. - No enactment. Do not start implementing the plan under discussion. The skill ends at the confirmed summary; acting on it is a separate, user-initiated step.
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.
- 6d ago First seen · 75 lines · 112 tokens per session scan A 7b722112c476
amadeus-grilling is a skill published in the GitHub repository amadeus-dlc/amadeus (8 stars, last pushed 20d ago), licensed Apache-2.0. It adds 112 tokens to every session and 915 once invoked, about $0.0006 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 2026-09-03.
Other skills, from other repositories
cocotrace
Build and query the SHA-256 requirements-to-implementation traceability graph across lifecycle artifacts.
cocoaudit
View and export the append-only session audit trail for regulated environments.
Changelog Test Mapper
Map changelog entries and release notes to affected test cases, ensuring every user-facing change has corresponding test coverage verification.
derive-nfr-from-driver
Derives measurable NFRs from a parent business driver.
derive-security-from-risk
Derives security requirements from a named threat.
traceability-gate
Enforces parent-link lineage across requirement layers.