Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add e1024kb/wise-claude/plugin install wiseWrote 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/e1024kb/wise-claude/wise-grill)<a href="https://agentmods.dev/skills/e1024kb/wise-claude/wise-grill"><img src="https://agentmods.dev/badge/skills/e1024kb/wise-claude/wise-grill.svg" alt="Measured on agentmods" 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 | $0.00268 | $0.03838 |
| Opus 5 | $0.00134 | $0.01919 |
| Sonnet 5 | $0.00054 | $0.00768 |
| Haiku 4.5 | $0.00027 | $0.00384 |
Grade A, and why
wise-grill 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 7d 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 — 308 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/wise-grill — understand an underspecified subject, then plan it, answer it, or grill the right people
Why this skill exists
The reasoning behind a piece of work rarely lives where the work is
stated. "Fix the export timeout" may mean "raise a config value" —
decided in a 40-message Slack thread the ticket never links; a spec
page may assume vocabulary only its author's team shares; a one-line
prompt may hide three unstated decisions. /wise-grill is the step
before anything is built on such input: it classifies what it was
given, researches everything reachable, states what the subject
actually means, and forks — a plan when the evidence supports one, a
researched answer when the input was a question, or a blueprint whose
questions are addressed to whoever can close each gap (a specific
person from the evidence, or the user when nobody else can answer).
Facts get researched; only decisions get asked.
Arguments
$ARGUMENTS — the subject plus optional guidance:
- A URL or bare ticket id first (
https://acme.atlassian.net/browse/PROJ-1,PROJ-123,#678, a Confluence / Notion / Google-Docs link, a path to an existingBLUEPRINT-*.md) — everything after the first whitespace is free-form guidance threaded through the whole pass (focus areas, constraints, known answers — e.g./wise-grill PROJ-1 the reporter told me it only affects the CSV export). - Anything else — the entire
$ARGUMENTSstring is a free-form prompt or question; there is no separate guidance tail (fold any embedded constraints in as part of the subject).
Procedure
1. Parse, validate, classify
If $ARGUMENTS is empty, ask one AskUserQuestion for the
subject, then proceed. If the first token is a placeholder
(<ticket>, $VAR, {id}, TODO, ..., ? alone), stop with a
clear error — provided-but-invalid is an error, never a re-prompt.
Classify the input into exactly one input_type:
input_type |
Recognise by | Subject / guidance split |
|---|---|---|
blueprint |
first token is a path to an existing BLUEPRINT-*.md, or docs/plans/BLUEPRINT-<ref>.md already exists for the derived ref |
first token / rest |
ticket |
first token is a tracker URL (*.atlassian.net/browse, linear.app, github.com/*/issues, gitlab.com/*/-/issues, app.asana.com, …) or a bare id (ABC-123, #678) |
first token / rest |
doc |
first token is a non-tracker document URL — Confluence / Notion / Google Docs–Drive / SharePoint / a wiki or RFC page; WebSearch an unknown host to tell tracker from doc |
first token / rest |
question |
free-form text asking for understanding, not work — "how / why / what / where / should we", a trailing ?, "explain", "compare" |
whole string is the subject |
prompt |
free-form text describing work to do ("add rate limiting to the export API") — the catch-all when nothing above matches | whole string is the subject |
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.
- 7d ago First seen · 308 lines · 268 tokens per session scan A 3983de3a80b1
wise-grill is a skill published in the GitHub repository e1024kb/wise-claude (4 stars, last pushed today), licensed MIT. It adds 268 tokens to every session and 3,838 once invoked, about $0.0013 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
docs-drift
Documentation-drift audit for this plugin monorepo. Audit either each scoped plugin's unreleased changelog claims or its latest shipped release from the previous reachable plugin-name--v tag, then verify that plugin READMEs, docs, CLAUDE.md or AGENTS.md files, and root documentation tell the truth for the matching…
cco-tools
Show what tools actually cost in tokens — learned per-tool averages from observed results, replacing the hardcoded MCP/Agent guesses.
test-run
Run plugin test suites in this monorepo and report a concise pass/fail summary. Optional plugin slug arg; without arg, runs all plugins under plugins/.
github-pr-creation
Creates GitHub Pull Requests with automated validation and task tracking. Use when user wants to create PR, open pull request, submit for review, or check if ready for PR. Analyzes commits, validates task completion, generates Conventional Commits title and description, suggests labels. NOTE - for merging existing…
design-system-reference
Style guides and implementation rules for frontend design. Works with design-discovery agent which handles context gathering and VS-based style recommendations. Contains detailed style guides, anti-patterns, and implementation checklists.
magic-compact
Compact this Claude Code session.