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 shrimbly/project-wiki --skill wiki-researchgit clone --depth 1 https://github.com/shrimbly/project-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/shrimbly/project-wiki/wiki-research)<a href="https://agentmods.dev/skills/shrimbly/project-wiki/wiki-research"><img src="https://agentmods.dev/badge/skills/shrimbly/project-wiki/wiki-research/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/shrimbly/project-wiki/wiki-research"><img src="https://agentmods.dev/badge/skills/shrimbly/project-wiki/wiki-research.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 | $0.00091 | $0.01022 |
| Opus 5 | $0.00046 | $0.00511 |
| Sonnet 5 | $0.00018 | $0.00204 |
| Haiku 4.5 | $0.00009 | $0.00102 |
Grade A, and why
wiki-research 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.
How it starts
The opening of the file, as written. The whole thing — 93 lines — stays where its author put it; the contents beside it link to each section on GitHub.
wiki-research
Deepen the wiki through research the user steers. The defining feature of this skill is the menu: you propose concrete, scoped research options and the user picks. Don't run a sprawling investigation unprompted.
1. Survey the wiki for gaps
Read wiki/index.md, wiki/overview.md, and wiki/open-questions.md. Skim the
entity/concept pages relevant to what the user mentioned. Look for:
- Open questions that research could resolve.
- Thin pages — entities/concepts asserted but under-developed.
- Undefined terms used across pages but never pinned down.
- External references the project leans on — competitors, standards, prior art, libraries/APIs, domain background — that we've never actually studied.
- Unvalidated assumptions stated as fact with no source.
2. Build a research menu
Turn the gaps into 3–8 concrete research options. Each option is one investigable question, not a vague area. For each, note:
- the question (specific and answerable),
- why it matters / what it unblocks,
- where you'd look (web search/fetch, Context7 for library/API docs, a specific competitor or spec, an MCP source),
- the expected output (which
sources/capture and/orwiki/page it produces).
Present the full menu to the user as a numbered list in your message so they see everything.
3. Let the user choose (interactive)
Use AskUserQuestion (multiSelect) to let the user pick which options to run.
AskUserQuestion allows up to 4 options per question, so:
- If you have ≤4 candidates, offer them directly.
- If you have more, put the highest-value 3–4 in the picker and tell the user the rest are in your numbered list above — they can select "Other" and name any by number, or ask you to re-surface a different batch.
Respect the user's scoping. If they narrow or redirect, follow it. If they add a question you didn't list, take it.
4. Run the selected research
For each chosen option:
- Use the right tool: WebSearch to find, WebFetch to read, Context7 for current library/framework/API docs, the relevant MCP for domain sources. Pull primary sources where you can; note when something is second-hand or uncertain.
- Capture vs. synthesize — keep the layers separate:
- If the research yields durable external material worth preserving (a
competitor's model, a spec, a key article), write a faithful capture in
sources/(type: researchortype: web), with the URLs consulted in frontmatter. - Then synthesize into the wiki: a
wiki/references/<name>.mdpage for external/competitive captures, or update the relevant entity/concept/decision page with the findings. Cite the new source(s).
- If the research yields durable external material worth preserving (a
competitor's model, a spec, a key article), write a faithful capture in
- If the research answers an open question, fold the answer into the right
page and move the question to Resolved in
wiki/open-questions.mdwith a pointer. - If it raises new questions, add them to
wiki/open-questions.md.
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 · 93 lines · 91 tokens per session scan A b38e6ab18389
wiki-research is a skill published in the GitHub repository shrimbly/project-wiki (5 stars, last pushed 1mo ago), licensed MIT. It adds 91 tokens to every session and 1,022 once invoked, about $0.0005 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
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…