agentmods.dev

deep-research

deep-research is a skill for Claude Code, Codex from earlyaidopters/marks-pi-harness. It costs 65 tokens per session (1,054 once invoked), scanned A, original, MIT.

A structured process for researching questions that require many web searches and pages. It works in rounds, keeping a rewritten report, a store of verified facts, and a record of failed searches.

In plain words
What is it for?
Use it for deep research, comparisons, “find the best” questions, and investigations that need more than a couple of web sources.
Why use it?
It prevents long research sessions from becoming a pile of unverified notes or from treating earlier guesses as established facts.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

not rated 11repo +1 1mo ago A scan 65 tokens original MIT

Good fit Use it for deep research, comparisons, “find the best” questions, and investigations that need more than a couple of web sources.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/earlyaidopters/marks-pi-harness/deep-research
View source ↗ earlyaidopters/marks-pi-harness
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.

Any agent
npx skills add earlyaidopters/marks-pi-harness --skill deep-research
Clone the repo
git clone --depth 1 https://github.com/earlyaidopters/marks-pi-harness

Made for: Claude Code, Codex.

Wrote 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.

agentmods badge for deep-research

README.md
[![agentmods](https://agentmods.dev/badge/skills/earlyaidopters/marks-pi-harness/deep-research/github.svg)](https://agentmods.dev/skills/earlyaidopters/marks-pi-harness/deep-research)
Your own site
<a href="https://agentmods.dev/skills/earlyaidopters/marks-pi-harness/deep-research"><img src="https://agentmods.dev/badge/skills/earlyaidopters/marks-pi-harness/deep-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.

agentmods 80×15 button for deep-research

Your own site · 80×15
<a href="https://agentmods.dev/skills/earlyaidopters/marks-pi-harness/deep-research"><img src="https://agentmods.dev/badge/skills/earlyaidopters/marks-pi-harness/deep-research.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 65 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,054 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00065 $0.01054
Opus 5 $0.00032 $0.00527
Sonnet 5 $0.00013 $0.00211
Haiku 4.5 $0.00006 $0.00105

Measured 10d ago against content hash 547bb0c75079, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, 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 10d 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.

skills/deep-research/SKILL.md · 80 lines

How it starts

The opening of the file, as written. The whole thing — 80 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Deep research (rounds + facts store)

Raw back-and-forth history poisons small models: by page 6 you have lost page 1, and your own earlier speculation starts reading like established fact. This skill replaces accumulation with ROUNDS. Scratch thinking is discarded every round; the only things that survive are the REPORT (rewritten each round), the FACTS STORE (facts_add/facts_recall), and the LEDGER of what failed.

Before round 1 — plan first (this turn decides the whole trajectory)

  1. facts_recall with the core question keywords. You may already know part of the answer from an earlier session.
  2. Check the question for keyword traps BEFORE searching:
    • Specific number in the topic ("42 year old") -> strip it unless removing it changes the meaning ("GPT-4" keep).
    • Tutorial phrasing ("how to use X") -> real discussions say "my X setup", "X in production". Reframe.
    • Generic single noun ("marketing") -> too broad; ask the user for a facet instead of running a doomed sweep.
    • A person's name that collides ("Kevin Rose") -> anchor EVERY query with a disambiguating entity ("kevin rose digg founder").
  3. Write the plan as your first Report (see format below): what must be verified, which sub-questions, which sources. 2-5 DIFFERENT-intent queries, passed as an ARRAY to one web_search call. Never put temporal words ("recent", "2026", "latest") or meta words ("news", "updates") in a query — use the days parameter for recency.

Each round: Think -> Report -> Action

  • Think: reason freely about what the last observation means. This is scratch — it will NOT be carried forward, so never cite your own Think text as evidence later.

  • Report: rewrite it FROM SCRATCH, high density. Not an append. Format:

    QUESTION: <the user's question>
VERIFIED (from fetched content, with source): ...
UNVERIFIED LEADS (from snippets only): ...
OPEN GAPS: what the answer still needs
FAILED: approaches/sources tried that produced nothing (never retry these)
NEXT: the single best next action and why

Read the full file on GitHub · 80 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. 10d ago First seen · 80 lines · 65 tokens per session scan A 547bb0c75079

Subscribe to this mod's changes

deep-research is a skill published in the GitHub repository earlyaidopters/marks-pi-harness (11 stars, last pushed 1mo ago), licensed MIT. It adds 65 tokens to every session and 1,054 once invoked, about $0.0003 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.

Research
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

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…

microsoft/ai-agents-for-beginners · 200 tokens

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…

vercel/next.js · 95 tokens

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…

vercel/next.js · 83 tokens

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…

vercel/next.js · 170 tokens

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…

vercel/next.js · 103 tokens