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 cskwork/grill-with-docs-html --skill skillgit clone --depth 1 https://github.com/cskwork/grill-with-docs-htmlWrote 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/cskwork/grill-with-docs-html/skill)<a href="https://agentmods.dev/skills/cskwork/grill-with-docs-html/skill"><img src="https://agentmods.dev/badge/skills/cskwork/grill-with-docs-html/skill/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/cskwork/grill-with-docs-html/skill"><img src="https://agentmods.dev/badge/skills/cskwork/grill-with-docs-html/skill.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 | $0.00061 | $0.01242 |
| Opus 5.5 | $0.00024 | $0.00497 |
| Sonnet 5.5 | $0.00012 | $0.00248 |
| Haiku 4.5 | $0.00006 | $0.00124 |
Grade A, and why
grill-with-docs-html 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 1mo 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.
Run the grilling interview exactly as grill-with-docs does — Socratic, one question at a time, recommended answer each time, codebase exploration over speculation, inline CONTEXT.md / ADR updates as decisions crystallise.
Default add-on (this skill): once the interview has produced 3+ structured decisions (typically D1.x / D2.x style grouped under candidates), export the full decision tree as a self-contained HTML form so the user can review the recommendations side-by-side, override any choice, write notes, and copy a final decision text back into chat.
Order of operations:
- Run the interview when needed (see
../grill-with-docs/SKILL.md); use existing agreed decisions directly when the user requests export only. - As decisions land, keep a mental ledger of: candidate id, decision id, options, trade-offs, recommended option, the "why" reasoning, side effects.
- Draft Before/After Mermaid diagrams for each candidate (current shape vs. deepened shape) — read
MERMAID-SAFE-SUBSET.mdbefore writing the first diagram. - Validate every Mermaid block with mmdc BEFORE writing the HTML. Read
VALIDATION.mdfor the extract-and-render snippet and run it — an unrendered block shows the user a syntax error instead of a diagram. - Write the HTML using the template in
HTML-FORM.mdto$TMPDIR/grill-decisions-<YYYYMMDD-HHMMSS>.html(fallback/tmp), then open it (see File location). - Return the form for review. Pasted choices record decisions; implement only when the user also authorizes implementation, and do not re-ask for authorization already given for the unchanged scope.
Reuses
This skill reuses the entirety of grill-with-docs:
- Domain awareness (CONTEXT.md / CONTEXT-MAP.md discovery, lazy creation) — see
../grill-with-docs/SKILL.md - Glossary challenges, fuzzy-term sharpening, scenario stress-tests
- Inline
CONTEXT.mdupdates via../grill-with-docs/CONTEXT-FORMAT.md - ADR offers via
../grill-with-docs/ADR-FORMAT.md(only when hard-to-reverse + surprising + real trade-off)
Read those files for the interview behaviour; this skill's bundled files only cover the HTML export layer. For the interview, resolve the installed grill-with-docs skill path; do not assume it is beside a symlink target. If unavailable, report the missing interview dependency. Export-only requests can use the supplied decisions and bundled HTML guidance without that dependency.
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 1mo ago Changed · -23 tokens per session 2160d8327b70
- 1mo ago First seen · 75 lines · 84 tokens per session scan A 84ea4452b5fd
grill-with-docs-html is a skill published in the GitHub repository cskwork/grill-with-docs-html (1 stars, last pushed 1mo ago), licensed MIT. It adds 61 tokens to every session and 1,242 once invoked, about $0.0002 per session on Opus 5.5. 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
declauding
Load at the start of any task whose deliverable is prose another person will read — a PR description, commit message, README, doc, postmortem, blog post, issue or review comment, report, release note or essay — before drafting it, whether the result is pushed, posted, published or saved to a file, without being asked.…
generative-thinking
Break out of a locked problem frame by picking one disciplined move — reframe, provocation (Po), random stimulus, SCAMPER, inversion, structured analogy, constraint play, or family traversal — and committing to it before evaluating. Use when stuck, when options feel narrow or obvious, when iterations produce…
swarming-to-green
Runs a long swarm of hundreds to thousands of cheap headless coding workers that drive a large failing-test backlog to green. Failures are clustered by signature, each worker fixes one cluster in its own git worktree, a serial no-regression merge gate lands the fixes, the loop resumes across restarts for hours, and it…
orchestrating-chat-agents
Fans Claude work out over the Anthropic API from surfaces with no native subagents, chiefly claude.ai chat and projects, with parallel calls with a shared cached context, streaming, long-lived conversation threads with cached history, and named agent pools; also delegates to Gemini through the Cloudflare AI Gateway on…
browsing-bluesky
Browse Bluesky content via API and firehose - search posts, fetch user activity, sample trending topics, read feeds and lists, analyze and categorize accounts. Supports authenticated access for personalized feeds. Use for Bluesky research, user monitoring, trend analysis, feed reading, firehose sampling, account…
hallucinating-labels
Assign items to a CLOSED label vocabulary that is too large to put in a prompt — product taxonomies, category hierarchies, tag vocabularies, routing tables, ICD/SIC-style code lists. A cheap model writes the label it thinks the vocabulary would use, and an embedder snaps that writing onto the nearest legal value, so…