Use when the user asks to research a topic and produce a thorough sourced report. Not for remote, credential, publish, deploy, or irreversible changes.
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# Deep research
## Contract
| Field | Bound contract |
|---|---|
| Trigger | User asks to research a topic and produce a thorough sourced report. |
| Authority | Reversible local: writes only ./research/{type}-{topic}-{YYYY-MM-DD}.md (or the optional output-path override) and an optional PDF beside it; rollback is version control or undo. No remote mutation. Spawns 3-20 read-only sub-agents. |
| Side effect | Writes ./research/{type}-{topic}-{YYYY-MM-DD}.md; optionally exports a PDF beside it. Sub-agents perform read-only web search and page fetch. Output path can be overridden with the optional output-path input. |
| Done | A prose-first cited Markdown report exists at the output path, containing key findings, strategic recommendations, and risks/uncertainties, with every claim cited. |
## Inputs
- Topic and research type, supplied by the user or inferred from the prompt. Research types: market, domain, technical, competitive, product, academic, person/org, financial, legal, trend, community. If none fit, infer a type and design a matching axis breakdown.
- Specific questions or goals the research must answer (optional for well-scoped prompts; required for vague prompts).
…
The line at the top of this page pins the label git:20260912.5c5b759, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
GET https://markdownregistry.com/api/v1/artifacts/art_3swi2svjdiyhhyy7
GET https://markdownregistry.com/api/v1/resolve?ref=outlinedriven/outline-driven-development/deep-research
GET https://markdownregistry.com/api/v1/blob/3f5ded0c386bbced3b2288e95e5ae5274dc9e085ee70473083204ecdc6d7f2c8
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
outlinedriven/outline-driven-development · .devin/skills/abi-and-calling-conventions/SKILL.md · Use when explaining System V AMD64, ARM AAPCS, RISC-V psABI, stack frames, variadic calls, or FFI register rules. Not…
outlinedriven/outline-driven-development · .devin/skills/abstract-well/SKILL.md · Use when the user wants to classify abstractions as useful, bad, or busy and keep one shallow level. Not for tasks…
outlinedriven/outline-driven-development · .devin/skills/adc-dac-baremetal/SKILL.md · Use when configuring ADC sampling time, DMA-driven ADC, calibration, or DAC channel setup on bare-metal MCUs. Not for…
outlinedriven/outline-driven-development · .devin/skills/address-sanitizer/SKILL.md · Use when building or running native code under AddressSanitizer, interpreting an existing ASan report, or debugging a…
outlinedriven/outline-driven-development · .devin/skills/af-xdp/SKILL.md · Use when creating AF_XDP sockets, configuring UMEM and XSK rings, writing an XDP redirect program, or choosing copy…
outlinedriven/outline-driven-development · .devin/skills/agent-environment-retrospective/SKILL.md · Use when a completed session needs an agent-environment retrospective. Not for an engineering retrospective from…
outlinedriven/outline-driven-development · .devin/skills/agent-surface-forge/SKILL.md · Use when asked to audit or repair agent surfaces (plugins, agents, skills, CLAUDE.md/AGENTS.md, docs, prompts…
outlinedriven/outline-driven-development · .devin/skills/agent-transcript/SKILL.md · Use when a redacted, trimmed agent transcript must be appended to a GitHub PR or issue body, with human approval and…
outlinedriven/outline-driven-development · .devin/skills/agents-md/SKILL.md · Use when a repo needs agent setup, AGENTS.md added or made lean, CLAUDE.md audited, or agent instructions scored or…
outlinedriven/outline-driven-development · .devin/skills/ai-collab-protocols/SKILL.md · Use when the user describes an AI workflow gap or uses an ambiguous cross-session reference such as ''the PR Bob…
outlinedriven/outline-driven-development · .devin/skills/all-scenarios-storm/SKILL.md · Use when a user wants to enumerate plausible designs, configurations, scenarios, and paths and diagram the field before…
outlinedriven/outline-driven-development · .devin/skills/analysis-artifacts/SKILL.md · Use when the user requests a deep dive, exploratory analysis, or data analysis on BigQuery. Not for credential…
affaan-m/ecc · .agents/skills/deep-research/SKILL.md · Multi-source deep research using firecrawl and exa MCPs. Searches the web, synthesizes findings, and delivers cited…
affaan-m/ecc · .kiro/skills/deep-research/SKILL.md · Multi-source deep research using firecrawl and exa MCPs. Searches the web, synthesizes findings, and delivers cited…
bytedance/deer-flow · skills/public/deep-research/SKILL.md · Use for substantial investigations needing current evidence, source comparison, or synthesis across several dimensions…
imbad0202/academic-research-skills · deep-research/SKILL.md · Universal deep research agent team. 13-agent pipeline for rigorous academic research on any topic. 8 modes: full…
thu-maic/openmaic · skills/agent-runtime/deep-research/SKILL.md · Courses whose content rests on current, external or real-world facts that must be verified against live sources before…
genspark-ai/genoffice · skills/recommended/deep-research/SKILL.md · Run multi-round searches with cross-checking and produce a research report with cited sources, delivered as a Word…