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 SaikaAco/saika-hermes-skills --skill bounded-grillinggit clone --depth 1 https://github.com/SaikaAco/saika-hermes-skillsWrote 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/saikaaco/saika-hermes-skills/bounded-grilling)<a href="https://agentmods.dev/skills/saikaaco/saika-hermes-skills/bounded-grilling"><img src="https://agentmods.dev/badge/skills/saikaaco/saika-hermes-skills/bounded-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/saikaaco/saika-hermes-skills/bounded-grilling"><img src="https://agentmods.dev/badge/skills/saikaaco/saika-hermes-skills/bounded-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 | $0.00062 | $0.02407 |
| Opus 5 | $0.00031 | $0.01203 |
| Sonnet 5 | $0.00012 | $0.00481 |
| Haiku 4.5 | $0.00006 | $0.00241 |
Grade A, and why
bounded-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 11d 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 — 268 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Bounded Grilling
Overview
Bounded Grilling exposes and resolves material human decisions without turning an interview into an endless session or implementation. The human owns value judgments and authority decisions. The agent discovers facts, frames options, recommends a default, returns a traceable decision packet, and stops.
Invocation gate
Run only when:
- the user explicitly asks to be grilled, interviewed, or stress-tested; or
- the shared
/wfrouter invokes one genuine human-owned ambiguity under the user's standing ask-not-assume instruction; or - the user explicitly invokes Wayfinder charting, which authorizes one bounded opening-router call; or
- an existing, named Wayfinder HITL decision ticket invokes the skill.
Outside those cases, an agent may briefly offer Bounded Grilling when a route-changing decision is blocking progress, but must not begin until the user accepts. Complexity or ambiguity alone is not permission to start.
Authority boundary
- Ask the human only for decisions; investigate discoverable facts first.
- Never answer the human side of a HITL decision.
- Never implement, publish, install, configure, activate, or otherwise mutate the product because the interview ended.
- Bounded Grilling does not claim, edit, close, link, or create tracker items.
- The invoking Wayfinder workflow owns all authorized tracker persistence and map, fog, frontier, and dependency updates.
- Stop immediately on a user stop signal.
Choose one mode
Opening router
Use for a loose plan or destination before choosing the planning route.
- Resolve no more than three distinct route-changing human decisions.
- Stop earlier when remaining uncertainty is non-blocking.
- Return exactly one route:
DIRECT_SPEC— the route is clear enough for local specification;WAYFINDER— material multi-session fog or dependent decisions remain;BLOCKED— required evidence or decision authority is unavailable.
- Return the compact decision packet and stop.
- Do not create a specification, Wayfinder map, task, or other artifact.
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.
- 11d ago First seen · 268 lines · 62 tokens per session scan A 04c4dbdca994
bounded-grilling is a skill published in the GitHub repository SaikaAco/saika-hermes-skills (2 stars, last pushed 12d ago), licensed MIT. It adds 62 tokens to every session and 2,407 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 2026-08-31.
Other skills, from other repositories
hunt-auth-bypass
Hunting skill for auth bypass vulnerabilities. Built from 12 public bug bounty reports across SAML XSW / parser-differential (GitHub Enterprise CVE-2025-25291/25292), SAML signature stripping (Uber, Rocket.Chat, samlify CVE-2025-47949), SAML domain enforcement bypass via control characters (HackerOne 2024)…
m365-entra-attack
Microsoft 365 / Entra ID red-team attack chain — current 2026 reality. AADSTS code reference, user enumeration vectors (with hardening status), Smart Lockout math, Conditional Access bypass options, ROPC + SAML SSO browser flow, Burp/Playwright templates. Built from authorized red-team work where ROPC spray surfaced…
password-spray-methodology
End-to-end password spray playbook. User enumeration, lockout detection, password pattern generation, spray execution across all protocols, error code differentials, and engagement discipline. Unifies M365/Entra, Okta, Exchange, Kerberos, SharePoint, XMLRPC, OIDC, and AD SMB/WinRM spraying into one methodology.
cache-attack
Poison CDN cache or deceive when X-Cache header is detected.
hunt-django
Hunt Django-specific vulnerabilities: DRF permission gaps, ORM injection, and admin exploitation.
tianya-gods-team
A decision-making system in which 20 fictional specialist viewpoints analyze one question in parallel before a coordinating AI combines them. It covers areas such as history, economics, relationships, technology, mysteries, and culture.