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 qhkm/zeptoclaw --skill deep-researchgit clone --depth 1 https://github.com/qhkm/zeptoclawWrote 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/qhkm/zeptoclaw/deep-research)<a href="https://agentmods.dev/skills/qhkm/zeptoclaw/deep-research"><img src="https://agentmods.dev/badge/skills/qhkm/zeptoclaw/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.
<a href="https://agentmods.dev/skills/qhkm/zeptoclaw/deep-research"><img src="https://agentmods.dev/badge/skills/qhkm/zeptoclaw/deep-research.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00015 | $0.00766 |
| Opus 5.5 | $0.00006 | $0.00306 |
| Sonnet 5.5 | $0.00003 | $0.00153 |
| Haiku 4.5 | $0.00002 | $0.00077 |
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 2mo 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 — 85 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Deep Research Skill
Use this skill when the user asks you to research a topic in depth, produce a report, or investigate something thoroughly. A single search query is NEVER enough for real research.
Methodology
Phase 1: Broad Exploration (3-5 searches)
Start wide. Use web_search with varied phrasings to map the territory:
- Search the main topic from different angles
- Identify subtopics, key players, and perspectives
- Note which areas have the most information vs gaps
Example: For "state of AI agents in 2026":
web_search("AI agent frameworks 2026")web_search("autonomous AI agents market landscape")web_search("AI agent orchestration tools comparison")
Phase 2: Deep Dive (2-3 searches + fetches per subtopic)
For each important subtopic from Phase 1:
- Use targeted, precise search queries
- Try multiple phrasings for the same concept
- Use
web_fetchto read full articles (not just search snippets) - Follow references and links mentioned in sources
Example:
web_search("LangGraph vs CrewAI vs AutoGen benchmark 2026")web_fetch("https://example.com/detailed-comparison-article")
Phase 3: Diversity & Validation
Ensure you have coverage across these 6 types of information:
- Facts & data — statistics, market size, benchmarks, numbers
- Real-world examples — case studies, actual implementations, user stories
- Expert opinions — interviews, commentary, analysis from known figures
- Trends & predictions — current year developments, future directions
- Comparisons — alternatives, trade-offs, vs analyses
- Challenges & criticisms — problems, limitations, balanced critique
Search specifically for any category you're missing.
Phase 4: Synthesis Check
Before writing your final answer, verify:
- Did I cover 3-5 different angles?
- Did I read important sources in full (not just snippets)?
- Do I have concrete data AND real examples AND expert perspectives?
- Did I include both positives and challenges?
- Are my sources current and authoritative?
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.
- 2mo ago First seen · 85 lines · 15 tokens per session scan A a3d3f9713167
deep-research is a skill published in the GitHub repository qhkm/zeptoclaw (654 stars, last pushed 7d ago), licensed Apache-2.0. It adds 15 tokens to every session and 766 once invoked, about $0.0001 per session on Opus 5.5. 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 2026-08-30.
deep-research compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| deep-researchqhkm/zeptoclaw | Skill | 654 | 7d ago | Apache-2.0 | A | 15 | Claude Code, Codex |
| anarlogfastrepl/anarlog | Skill | 9.8k | yesterday | MIT | A | 50 | Claude Code |
| llama-cppmoltis-org/moltis | Skill | 2.9k | 14d ago | MIT | A | 91 | Claude Code, Codex |
| app-buildervellum-ai/vellum-assistant | Skill | 1.5k | 2d ago | MIT | A | 162 | Claude Code, Codex |
| gmailvellum-ai/vellum-assistant | Skill | 1.5k | 2d ago | MIT | A | 20 | Claude Code, Codex |
| guardian-verify-setupvellum-ai/vellum-assistant | Skill | 1.5k | 2d ago | MIT | A | 26 | Claude Code, Codex |
| systematic-debuggingmoltis-org/moltis | Skill | 2.9k | 14d ago | MIT | A | 33 | Claude Code, Codex |