deep-research

A workflow for conducting multi-step research: plan the question, search sources, extract and assess information, combine the findings, and save the result.

In plain words
What is it for?
Use it to produce research reports from web pages, academic sources, local knowledge, and cached results.
Why use it?
It organizes scattered research work into a repeatable process and keeps track of source quality and missing information.

Skill for Claude CodeCodex

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/mistakeknot/interdeep/deep-research
Any agent
npx skills add mistakeknot/interdeep --skill deep-research
Clone the repo
git clone --depth 1 https://github.com/mistakeknot/interdeep

Made for: Claude Code, Codex.

Per session 19 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,232 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 $0.00019 $0.01232
Opus 5 $0.00010 $0.00616
Sonnet 5 $0.00004 $0.00246
Haiku 4.5 $0.00002 $0.00123

Measured yesterday against content hash 9f1a8b8a70d7, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

deep-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 yesterday.

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.

skills/deep-research/SKILL.md · 129 lines

How it starts

The opening of the file, as written. The whole thing — 129 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/interdeep:deep-research

Orchestrate a deep research session: decompose a query, search multiple sources, extract and evaluate content, synthesize findings, and persist the result.

Protocol

Execute these five phases in order. Adapt depth based on the mode (see Depth Modes below).

Phase 1: Orient

  1. Use the research-planner agent to decompose the query into sub-queries with source routing.
  2. Review the planner output: query_type, sub_queries (each with query, sources, priority), depth_recommendation, rationale.
  3. If the user did not specify a depth mode, use the planner's depth_recommendation.

Phase 2: Search

For each sub-query, dispatch searches to the routed sources:

  • Web search — call interject_search (interject plugin) or web_search_exa (interflux/exa plugin):
    mcp tool: interject_search
    args: { "query": "<sub_query>", "source": "exa", "max_results": 10 }
    
  • Academic — call interject_search with "source": "arxiv".
  • Knowledge base — call interknow_qmd__search or interknow_qmd__vector_search (interknow plugin) for local knowledge.
  • Cached results — check intercache if available to avoid redundant fetches.

Collect URLs and snippets from all search results.

Phase 3: Extract

For each URL returned in Phase 2:

  1. Call extract_content (interdeep MCP tool) for single URLs:
    mcp tool: extract_content
    args: { "url": "<url>", "include_metadata": true }
    
  2. For batches (5+ URLs), use extract_batch:
    mcp tool: extract_batch
    args: { "urls": ["<url1>", "<url2>", ...], "max_concurrent": 5 }
    
  3. Use the source-evaluator agent on each extraction result to score relevance and credibility.
  4. Filter to sources where include_in_report is true.

Phase 4: Synthesize

  1. If interlens is available, call detect_thinking_gaps to identify blind spots:
    mcp tool: detect_thinking_gaps
    args: { "context": "<research summary so far>" }
    
  2. If gaps are found and depth mode is deep, run additional searches targeting the gaps.
  3. Pass evaluated findings to the report-compiler agent.
  4. Alternatively, call compile_report (interdeep MCP tool) for a structured markdown report:
    mcp tool: compile_report
    args: {
      "title": "<report title>",
      "query": "<original query>",
      "findings": [...],
      "sources": [...]
    }
    
  5. If intersynth is available, use it for additional synthesis passes.

Read the full file on GitHub · 129 lines

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. yesterday First seen · 129 lines · 19 tokens per session scan A 9f1a8b8a70d7

Subscribe to this mod's changes

deep-research is a skill published in the GitHub repository mistakeknot/interdeep (0 stars, last pushed 1mo ago), licensed MIT. It adds 19 tokens to every session and 1,232 once invoked, about $0.0001 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.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

brainstorming

You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.

obra/superpowers · 37 tokens

chat-pet-sprite-creation

Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.

microsoft/vscode · 53 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens

agent-host-chat-contributions

Build and review cross-cutting agent-host chat behavior through lifecycle contributions. Use when adding turn lifecycle side effects, prompt or context injection, restored-history transformation, protocol-action observation, or when reviewing changes that add code to AgentSideEffects or AgentService.

microsoft/vscode · 56 tokens

auto-perf-optimize

Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.

microsoft/vscode · 62 tokens