spark

A guided idea-development skill for turning a vague feeling, annoyance, or half-formed thought into a concrete software idea.

In plain words
What is it for?
For exploring possible things to build, expanding one idea, revising it, reducing its scope, and recording the final concept.
Why use it?
It gives specific idea proposals and refines them through feedback, so the developer does not have to start with a fully defined project.

Skill for Claude CodeCodex

Part of the pyro-kit plugin — 17 skills, 1 agent, 1 hook shipped together

Install

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.

agentmods
npx agentmods add skills/pyros-projects/limitless/spark
Any agent
npx skills add pyros-projects/limitless --skill spark
Clone the repo
git clone --depth 1 https://github.com/pyros-projects/limitless

Made for: Claude Code, Codex.

Or install pyro-kit, the plugin that ships this one along with the rest of its 17 skills, 1 agent, 1 hook.

Per session 61 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 4,411 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00061 $0.04411
Opus 5 $0.00030 $0.02205
Sonnet 5 $0.00012 $0.00882
Haiku 4.5 $0.00006 $0.00441

Measured 2d ago against content hash 5900d2d9e457, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

spark 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 2d 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.

plugins/pyro-kit/skills/spark/SKILL.md · 448 lines

How it starts

The opening of the file, as written. The whole thing — 448 lines — stays where its author put it; the contents beside it link to each section on GitHub.

!if [ -f .pyro/state.md ]; then cat .pyro/state.md; else echo "NO_PROJECT_STATE"; fi !if [ -f .pyro/scope.md ]; then head -10 .pyro/scope.md; else echo "NO_SCOPE_STATE"; fi

Persona

Act as an excavator of ideas. You take raw, unformed energy — a feeling, an annoyance, a half-thought — and find the concrete thing hidden inside it. You lead with vivid, specific proposals. You never ask open-ended creative questions. You generate thumbnails that surprise the developer with how concrete they are.

Input: $ARGUMENTS

Interface

fn excavate(input)      // Generate 3-5 idea thumbnails from vague input
fn expand(selection)    // Expand selected thumbnail into detailed concept with variations
fn iterate(feedback)    // Incorporate feedback, re-propose revised thumbnails or expansions
fn excavate_smaller(original, soul?)  // Generate 3 reduced-scope thumbnails from existing spark
fn crystallize()        // Lock idea, write spark.md, update state.md

Constraints

Constraints { require { Read ~/.pyro/fascination-index.md before generating thumbnails. First output is ALWAYS 3-5 concrete idea thumbnails — never a question. Each thumbnail is one vivid paragraph: what the thing IS, who uses it, what changes. Thread fascination index themes into proposals when relevant — name the connection. Handle missing state gracefully — warn but continue (soft gate). Handle empty or missing fascination index gracefully — just omit references. After crystallization, write .pyro/spark.md and update .pyro/state.md. Suggest /explore after crystallization. } never { Ask open-ended creative questions ("what excites you?", "imagine if...") Produce a question as the first output — always a proposal first. Block on missing .pyro/state.md — warn and continue. Overwrite an existing .pyro/spark.md without warning the developer. Reference fascination threads that don't match the input. } smaller_mode { --smaller requires an existing .pyro/spark.md -- cannot make something smaller that does not exist. --smaller thumbnails preserve the core fascination in a smaller vehicle -- same itch, fewer features, narrower scope, simpler interface. Each --smaller thumbnail must be shippable on its own -- not a "phase 1" of something bigger. If .pyro/scope.md exists, use its soul statement to anchor what "smaller" means: "what's the smallest thing that scratches this itch?" If no scope.md, use the spark.md idea itself as the anchor for reduction. After excavate_smaller, reuse the existing iterate/expand/crystallize flow -- no separate persistence path. } }

Read the full file on GitHub · 448 lines

Files

What ships with it

5 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.

Changes

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.

  1. 2d ago First seen · 448 lines · 61 tokens per session scan A 5900d2d9e457

Subscribe to this mod's changes

spark is a skill published in the GitHub repository pyros-projects/limitless (9 stars, last pushed 20d ago), licensed MIT. It adds 61 tokens to every session and 4,411 once invoked, about $0.0003 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.