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 agentmods add skills/jigripokri/poha/roastnpx skills add jigripokri/POHA --skill roastgit clone --depth 1 https://github.com/jigripokri/POHAWrote 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/jigripokri/poha/roast)<a href="https://agentmods.dev/skills/jigripokri/poha/roast"><img src="https://agentmods.dev/badge/skills/jigripokri/poha/roast.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 | $0.00023 | $0.00505 |
| Opus 5 | $0.00012 | $0.00253 |
| Sonnet 5 | $0.00005 | $0.00101 |
| Haiku 4.5 | $0.00002 | $0.00051 |
Grade A, and why
roast 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 4d 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.
What it actually says
/roast
User wants to deploy a roast in a group chat. You write it. They send it.
Recipe
-
Identify the target (a person, a recent message they sent, a running joke).
-
Read the last 30–50 messages in the group chat. Critical step — the roast has to match the group's existing tone and reference shared material.
-
Identify the roastable seam:
- A recent flex that backfired
- A predictable behavior pattern of the target
- A callback to an old joke at their expense
- A new piece of information that contradicts their stated position
- A comparison that punctures their self-image
-
Calibrate intensity to the group:
- Light: gentle ribbing, lots of love underneath
- Medium: pointed, but with a clear out (target can laugh)
- Hot: no quarter, but stays inside group norms
Default to Medium unless the group's recent tone is clearly Hot.
-
Generate 2–3 options. Different angles or escalation levels.
-
Show the options. Wait for "send X."
-
On approval, send via WhatsApp/Slack/wherever the group lives.
Roast rules (hard limits)
- No body-shaming. Ever.
- No real shots at the target's family unless they've made a self-deprecating reference first.
- No racism, sexism, homophobia, transphobia. Even when group "ironically" jokes that way.
- No quoting from carry-on.md. This skill cannot reference any private file.
- No bringing up the target's actual current pain (job loss, health, relationship). Roasts are for ambient material, not actual wounds.
- One landed roast > three forced ones. If nothing's good enough, say so. "Nothing roastable in this thread right now — wait for them to set something up."
Voice rules
Mirror the group's vocabulary, slang, and inside jokes. A good roast feels like the group wrote it, not an outsider.
If the user wants the AI's voice to come through, that's a different skill — this one is invisible.
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.
- 4d ago First seen · 50 lines · 23 tokens per session scan A 2167fc8605f9
roast is a skill published in the GitHub repository jigripokri/POHA (118 stars, last pushed 3mo ago), licensed MIT. It adds 23 tokens to every session and 505 once invoked, about $0.0001 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 2026-08-30.
Other skills, from other repositories
seedance
Generate AI videos using Volcengine Seedance model. Supports text-to-video (T2V), image-to-video (I2V), and audio-synced video generation. Use this skill when the user wants to create or generate videos.
music-search
Search cloud drives for downloadable music resources (songs, albums, lossless audio). Use this skill when the user wants to download a specific song or album. Do NOT use for general music information, lyrics, or recommendations.
p5js
Production pipeline for interactive and generative visual art using p5.js. Creates browser-based sketches, generative art, data visualizations, interactive experiences, 3D scenes, audio-reactive visuals, and motion graphics — exported as HTML, PNG, GIF, MP4, or SVG. Covers: 2D/3D rendering, noise and particle systems…
ascii-video
Production pipeline for ASCII art video — any format. Converts video/audio/images/generative input into colored ASCII character video output (MP4, GIF, image sequence). Covers: video-to-ASCII conversion, audio-reactive music visualizers, generative ASCII art animations, hybrid video+audio reactive, text/lyrics…
songwriting-and-ai-music
Songwriting craft, AI music generation prompts (Suno focus), parody/adaptation techniques, phonetic tricks, and lessons learned. These are tools and ideas, not rules. Break any of them when the art calls for it.
songsee
Generate spectrograms and audio feature visualizations (mel, chroma, MFCC, tempogram, etc.) from audio files via CLI. Useful for audio analysis, music production debugging, and visual documentation.