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 escoffier-labs/skillet --skill specialgit clone --depth 1 https://github.com/escoffier-labs/skilletWrote 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/escoffier-labs/skillet/special)<a href="https://agentmods.dev/skills/escoffier-labs/skillet/special"><img src="https://agentmods.dev/badge/skills/escoffier-labs/skillet/special/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/escoffier-labs/skillet/special"><img src="https://agentmods.dev/badge/skills/escoffier-labs/skillet/special.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 | $0.00083 | $0.01434 |
| Opus 5 | $0.00042 | $0.00717 |
| Sonnet 5 | $0.00017 | $0.00287 |
| Haiku 4.5 | $0.00008 | $0.00143 |
Grade A, and why
special 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 10d 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 — 96 lines — stays where its author put it; the contents beside it link to each section on GitHub.
special
The audit skills walk the line looking for what is broken. special walks the walk-in looking for what is possible. The chef sees what is fresh and plentiful on the shelves and proposes the dish worth adding to the board tonight, the one that uses what is already in the kitchen. A special is not invented out of nothing; it is the obvious next thing made from what you already have.
Core principle: every proposal is grounded in evidence already in the repo. If you cannot point to the signal in the walk-in, it is not a special, it is a daydream. No trend-chasing, no wishlist, no feature you cannot tie to something that already exists.
Read-only. special proposes; it never builds. Picking the dish and cooking it are later steps.
What this is not
- Not the audit trio. line-check, bug-hunt, and security-sweep find problems.
specialfinds opportunities. A bug is not a feature. - Not mise. mise designs an idea you already have.
specialfinds the idea worth designing. A chosen special flows into mise next.
Read the kitchen first
Before proposing anything, read what the project says it is and is not: README, CLAUDE.md / AGENTS.md, docs, declared roadmap, stated non-goals. A proposal that contradicts a declared non-goal is cut, or flagged as a deliberate challenge to it with that called out. Know the dish before you suggest the next course.
The signals (where specials come from)
Each lens reads the existing repo for evidence of an obvious next step. With parallel subagents, one finder per lens; otherwise sequential.
| Lens | The signal |
|---|---|
| Unfinished work | TODO/FIXME, stubbed functions, disabled tests, commented-out code that hints at intent, "coming soon" in docs |
| Asymmetry | Supports X for A but not B; read path with no write path; create/update with no delete; one integration where the shape implies two |
| Adjacent capability | One small step from something that exists: you parse it, so you could validate it; you have the data, so you could export it |
| Friction | Manual runbook steps that beg automation, repeated operator actions, a known-limitations or FAQ section describing recurring pain |
| Ecosystem fit | A capability comparable tools in this exact niche have that this lacks AND that serves the project's stated goals, not a fashionable add-on |
What ships with it
2 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.
- 10d ago First seen · 96 lines · 83 tokens per session scan A 399bbe9b1796
special is a skill published in the GitHub repository escoffier-labs/skillet (4 stars, last pushed 9d ago), licensed MIT. It adds 83 tokens to every session and 1,434 once invoked, about $0.0004 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
assess-interview-candidate
A structured hiring-assessment workflow that turns a candidate's résumé and a job description into reviewable evidence, interview questions, and an offline HTML report for interviewers.
task-clarifier
Deep need-clarification skill. Use only when the user explicitly invokes $task-clarifier. Once activated, keep asking until all three goals are met: the user fully understands their own needs, the AI fully understands the user's needs, and the user confirms the AI's understanding is correct. Do not intervene in task…
user-profile-keeper
Local user-profile maintenance skill for Codex, Claude Code, OpenClaw, OpenCode, and other agent harnesses. Use only when the user explicitly invokes $user-profile-keeper to create, initialize, update, query, correct, delete, export, or audit a local persistent user profile. It can extract durable collaboration…
run-history-skill-builder
Turn a completed task, browser flow, artifact pipeline, failure-recovery trace, or repeatedly refined workflow into a new reusable skill package or a reviewed skill-design plan. Use when the user asks to make a new skill from real run history, extract a reusable workflow from conversation/logs/files, summarize lessons…
run-history-skill-upgrader
Use real run evidence, validation failures, source drift, platform drift, and user feedback to plan and, only after explicit approval, apply structural upgrades to an existing skill. Use when the user asks to improve an existing skill from recent runs, recurring failures, outdated sources, excessive bloat, changed…
session-handoff-prompt
Create a concise continuation prompt that a fresh agent session can paste in to resume a long or degraded session. Use when the user asks for a handoff prompt, restart prompt, continuation prompt, context transfer, fresh-session resume, or a compact summary for opening a new session. Do not use for ordinary summaries…