Borrowing it
Nothing to install: this file belongs to hlalljie/agent-workflow-presets. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/hlalljie/agent-workflow-presets/main/.cursor/skills/research/SKILL.mdgit clone --depth 1 https://github.com/hlalljie/agent-workflow-presetsWrote 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/hlalljie/agent-workflow-presets/research)<a href="https://agentmods.dev/skills/hlalljie/agent-workflow-presets/research"><img src="https://agentmods.dev/badge/skills/hlalljie/agent-workflow-presets/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/hlalljie/agent-workflow-presets/research"><img src="https://agentmods.dev/badge/skills/hlalljie/agent-workflow-presets/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 | $0.00036 | $0.00787 |
| Opus 5 | $0.00018 | $0.00394 |
| Sonnet 5 | $0.00007 | $0.00157 |
| Haiku 4.5 | $0.00004 | $0.00079 |
Grade A, and why
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 8d 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 — 55 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Research
When to use
Invoked explicitly when a decision needs information gathering. Always spawn a subagent to execute this skill using [MODEL:researcher]. Not every question needs a full research doc — if a quick web search answers it, just answer in chat. Use this skill when the topic is complex, has multiple options to weigh, or will be referenced later.
Research produces a standalone reference doc. It does not contain status markers, links to open questions, or links to plans. Other docs link to it; it links to nothing upstream.
File location: always write research docs to .cursor/plans/<plan-name>/research/<topic>.md.
Process
- Clarify what the user is asking. A research doc can cover multiple related questions if they naturally group together — it is organized around a topic, not rigidly one-question-per-file.
- Search broadly (multiple queries, different angles).
- Fetch and read promising sources before citing them. If a source looks relevant but cannot be read (paywall, video, etc.), move on unless nothing else covers that ground — then link it with a note.
- Discard junk. A source earns its place by directly informing the user's question. Do not pad with tangentially related links.
- Write the doc.
- Tell the user what you found.
Search strategy
- Multiple web searches with different framing. Include the current year in searches — information must be current. Do not cite old sources unless nothing has changed in that area (rare for tech topics).
- Fetch and read article full text before citing. Summarize what the article argues, not what the title says.
- Videos: include when the topic is better explained visually or the user asks for videos. Search for short (<20 min), recent, popular results. Provide link, duration, channel. Do not over-explain that you cannot watch them — the user knows.
- If web search is not surfacing YouTube results, try
youtube <topic> <year>. If that fails, give the user search terms. - Quality over quantity. 2 good sources beat 10 unvetted ones.
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.
- 8d ago First seen · 55 lines · 36 tokens per session scan A 622b3a30de44
research is a skill published in the GitHub repository hlalljie/agent-workflow-presets (2 stars, last pushed 2mo ago), licensed MIT. It adds 36 tokens to every session and 787 once invoked, about $0.0002 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
audit
Eight audit modes (codebase — owns "audit the codebase" — docs/process, performance, threat-model, motion, SEO, debt). Triggers "nuclear review", "whole codebase review", "adversarial audit", "fable audit", "correctness audit", "audit the docs", "doc drift", "process audit", "perf audit", "why is it slow", "bundle…
autoresearch
Autonomous skill-prompt optimization — Karpathy-style mutate/score/keep loop on SKILL.md. Triggers "autoresearch", "optimize skill", "tune", "evolve" a skill, "prompt optimization".
design-tokens
Generate WCAG-checked type, color, spacing, and dark-mode tokens, or consolidate Tailwind v4 tokens with identical rendering. Triggers "design tokens", "type scale", "color palette", "theme setup", "reduce tokens", "dedupe tokens", "consolidate tokens".
lighthouse
Lighthouse audit + improvement loop until targets met. Triggers "lighthouse", "page speed", "improve scores", "LCP", "CLS", "INP", "core web vitals". Repo-wide perf audits ("perf audit") go to /audit performance mode instead.
orchestrate
Multi-agent coordination across plan/implement/test/review; fan-out for 3+ workstreams. Triggers "coordinate", "orchestrate", "parallel agents", "fan out", "split work", refactor, "overnight", "autonomous task", "marathon".
plan-ceo-review
CEO/founder-mode plan review — stress-tests a plan through a product lens before committing. Triggers "ceo review", "founder review", "product review", "should we even build this", or before a large architectural commitment.