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 schub-tech/llm-ideation-wiki --skill find-gapsgit clone --depth 1 https://github.com/schub-tech/llm-ideation-wikiWrote 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/schub-tech/llm-ideation-wiki/find-gaps)<a href="https://agentmods.dev/skills/schub-tech/llm-ideation-wiki/find-gaps"><img src="https://agentmods.dev/badge/skills/schub-tech/llm-ideation-wiki/find-gaps/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/schub-tech/llm-ideation-wiki/find-gaps"><img src="https://agentmods.dev/badge/skills/schub-tech/llm-ideation-wiki/find-gaps.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 | $0.00114 | $0.00616 |
| Opus 5 | $0.00057 | $0.00308 |
| Sonnet 5 | $0.00023 | $0.00123 |
| Haiku 4.5 | $0.00011 | $0.00062 |
Grade A, and why
find-gaps 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 9d 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
Convert the wiki's existing open questions into a concrete research plan. Don't invent new questions — start from what the idea overview already flags as unvalidated, then group, prioritize, and propose a strategy for each.
Read the wiki first. Start with wiki/index.md, then the idea overview, then its deep-dive notes. Pull every "Still to validate" line, every "What we do not know" entry, and every numbered move under "Next moves." Skip questions the wiki already answers — those are noise.
Cluster each gap by research strategy:
- Ask the user. Things only the user knows — domain experience, network, prior conversations, direct buyer access, founder-fit detail.
- Web research. Things that should be in public sources — competitor pricing, regulatory deadlines, named partners, market sizing, dated triggers.
- Hybrid. Things that need a user hint to find the right web target ("is there a Texas DIR equivalent in your actual market?").
Propose the plan as a numbered list. For each item, name the question, the strategy, the wiki section it would feed, and rough effort (one search vs. deep dive). Lead with questions that would flip the verdict — fatal-flaw checks first, decoration last. If a question is decorative (would not change the verdict either way), say so and recommend dropping it.
Confirm scope with the user before executing:
- Which questions to pursue this round (don't over-batch).
- Time range for web research (e.g., "since 2024", "all time").
- Depth (one search → one note vs. multiple search rounds → comprehensive note).
- Whether to run web items in parallel via
/research-deep.
Hand off. For web items, invoke /research-deep with the confirmed questions. For user-question items, surface them inline one at a time and capture the answers — short answers go directly into the relevant Still to validate line in the Notion wiki page; longer answers may warrant a new raw source page under Notion raw/<idea-slug>/.
Tie every proposed question to a specific wiki claim it would confirm or kill (e.g. "this would test the 60%-selector-drift bank-case figure"). Don't propose research the user could answer in two minutes. If a "Still to validate" line hedges, sharpen the question before researching it.
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.
- 9d ago First seen · 28 lines · 114 tokens per session scan A 0e1ee7c8ba8c
find-gaps is a skill published in the GitHub repository schub-tech/llm-ideation-wiki (11 stars, last pushed 1mo ago), licensed MIT. It adds 114 tokens to every session and 616 once invoked, about $0.0006 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
duckduckgo-search
Free keyless web, news, and image search via ddgs.
subagent-driven-development
Execute plans via delegatetask subagents (2-stage review).
mcporter
List, auth, and call MCP servers/tools from the terminal.
mem0-oss-to-platform
Plan and then execute a migration of a project from the mem0 open-source / self-hosted SDK (the local Memory class) to the mem0 Platform / hosted / managed SDK (the MemoryClient class). Use this whenever a developer wants to move, switch, or migrate their mem0 usage off OSS/self-hosted to the hosted API — e.g.…
deploy-docker-compose
Run the Omnigent server as a Docker compose stack (server + Postgres) on any Docker host — your laptop, a VPS, EC2 by hand, or as the base layer of any container-platform deploy. Invoke when the user wants to build the image, bring up the compose stack, debug the stack on a host they already have, or extend the stack…
mapping-to-snomed
Maps clinical concept spans extracted by OpenMed to SNOMED CT concepts through a USER-SUPPLIED terminology server (the user's own Ontoserver, Snowstorm, or UMLS/UTS), never a bundled vocabulary. Use when the user wants to code findings, disorders, procedures, body structures, or substances to SNOMED CT, run an ECL…