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 rfxlamia/pocketto --skill pocket-pitchinggit clone --depth 1 https://github.com/rfxlamia/pockettoWrote 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/rfxlamia/pocketto/pocket-pitching)<a href="https://agentmods.dev/skills/rfxlamia/pocketto/pocket-pitching"><img src="https://agentmods.dev/badge/skills/rfxlamia/pocketto/pocket-pitching/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/rfxlamia/pocketto/pocket-pitching"><img src="https://agentmods.dev/badge/skills/rfxlamia/pocketto/pocket-pitching.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 | $0.00109 | $0.02519 |
| Opus 5.5 | $0.00044 | $0.01008 |
| Sonnet 5.5 | $0.00022 | $0.00504 |
| Haiku 4.5 | $0.00011 | $0.00252 |
Grade A, and why
pocket-pitching 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 15d 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 — 256 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Pocket Pitching
Pre-grinding problem exploration. Scans project context before making project-specific claims, picks 3-5 brainstorming methods based on problem type, runs them with independent advisor curation, investigates technical unknowns that could change the approach, then produces a pitch exploration doc with a problem statement and 2-3 approach directions.
Core principle: Diverge first, converge second. Never propose solutions until the problem is fully explored.
Start from the user's request. Use clear instructions, corrections, and approvals already given; do not ask the user to repeat them or select a numbered option. Ask only when an unanswered product decision or fact would materially change the exploration. Otherwise, inspect available context, proceed, and label any assumption that affects the result.
When to Use
Trigger when:
- Problem is vague, unformed, or needs exploration before spec-writing
- User says "pitch", "explore", "I have a rough idea", "help me think through this"
- Starting a new feature with no clear direction
Do NOT use:
- When problem is already clearly defined (use pocket-grinding)
- When a spec already exists (use pocket-grinding or pocket-planning)
- For "brainstorm" keyword — pocket-grinding owns that trigger
Boundary with pocket-grinding
| pocket-pitching | pocket-grinding |
|---|---|
| Explores the problem space | Specifies the solution |
| Produces: directions (sync vs async, new module vs extension) | Produces: concrete architecture + GWT scenarios |
| No GWT scenarios — just directional | Full BDD spec with acceptance criteria |
| 2-3 approach directions, no design | Design proposals with scenarios-validated tradeoffs |
When pitching output lands in docs/pocket/spec/, grinding reads it as INPUT context — not as a spec.
Operating Rules
- Use project evidence. Scan the project before making project-specific claims. If the user already gave the topic, begin exploring it; do not restart intake with a generic greeting.
- Confirm only new decisions. Restate the problem and continue when the summary preserves the user's meaning. Ask for confirmation only when your interpretation adds a consequential assumption or a product choice remains open. Treat the user's clear correction or approval as the answer.
- Get independent curation. After all method results are visible, call
advisor(). If it is unavailable or fails, use theadvisorsubagent fallback inreferences/advisor-brainstorm-protocol.md. Do not present your own synthesis as independent review. - Change approach when evidence stalls. For a technical unknown, try a distinct source or investigation mode when the first attempt does not answer it. A call count or retry limit is a cue to reassess, not a reason to stop. Record the evidence and carry any unresolved question into the brief and pocket-grinding.
- Honor an authorized handoff. If the user asks to continue to pocket-grinding, invoke it and pass the brief and open questions. If they have not chosen what to do next, present the choices once and wait.
What ships with it
6 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.
- 15d ago Changed · +1 tokens per session e743cb00b916
- 1mo ago First seen · 256 lines · 108 tokens per session scan A 27a3b4d4f216
pocket-pitching is a skill published in the GitHub repository rfxlamia/pocketto (6 stars, last pushed 7d ago), licensed MIT. It adds 109 tokens to every session and 2,519 once invoked, about $0.0004 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
adhd
Divergent ideation across model families — the same brainstorm, but each cognitive frame is handed to a different model (Claude sub-agent, OpenAI Codex, a cheap third via OpenCode) so the ideas diverge by frame and by model priors at once. Use for open design questions, architecture and API shape, naming, schema…
x-bug2rag
A knowledge-capture tool that turns reusable bug explanations into a local RAG collection, meaning a searchable store of text that an agent can retrieve later. It records the trigger, incorrect implementation, correct implementation, and observable difference.
fec-drawio-studio
An editable workflow for making technical diagrams in draw.io (also called diagrams.net), with the original .drawio file saved alongside exported images or documents. It covers architecture, database, UML, sequence, flow, machine-learning, and code-structure diagrams.
x-dev-rag-call
A local semantic search tool that finds relevant passages in specified Markdown or text files. It uses an embedding model, which represents meaning as numbers, to return the closest passages for a question or plan.
independent-review-loop
Independent review loop run before delivery in code work. An independent reviewer (Codex, or fresh Codex-style subagents) reads the whole PR against the approved outcome; the lead fixes in-scope functional findings and re-reviews until none remain. Its differentiator is independence and exhaustiveness — a reviewer…
resolve-dispute
Resolves stuck review findings via a put-up-or-concede exchange. Invoked by the facilitator when a finding survives arbitration without new evidence.