cbd-thc-research-workflow
CBD to THC conversion research workflow: academic literature gathering, catalyst screening, and safety boundaries for regulated chemistry research.
- License 许可
- MIT
- 所属仓库
- zapabob/Skills
- 最近更新
- 2026-10-26
- Stars
- ★ 2
- Forks
- 0
01 前言 Front Matter
- name
- cbd-thc-research-workflow
- author
- Hermes Agent
- license
- MIT
- version
- 0.1.0
- metadata
- { "hermes": { "tags": [ "chemistry", "cbd", "thc", "research", "regulated-substances", "OpenManus", "web-search" ], "category": "research", "related_skills": [ "regulated-chemistry-research-safety" ] } }
- platforms
- [ "windows", "linux", "macos" ]
- description
- CBD to THC conversion research workflow: academic literature gathering, catalyst screening, and safety boundaries for regulated chemistry research.
03 SKILL.md 正文
CBD to THC Conversion Research Workflow
This skill documents a safe, academic-focused research workflow for studying CBD-to-THC conversion, emphasizing literature gathering, catalyst screening, and strict safety boundaries. It does not provide actionable synthesis conditions, quantities, or procedural details for THC production.
When to Use
- Requests to review academic literature on CBD→THC conversion mechanisms
- Inquiry into catalyst screening (zeolites, acids, other materials) for qualitative comparison
- Assessment of isolation/purification methods as analytical concepts
- Framing research questions within regulated chemistry safety boundaries
- Redirecting users from actionable synthesis toward safe research scopes
Prerequisites
- Access to academic databases (PubMed, arXiv, Google Scholar) via OpenManus or web_search
- Awareness that Japanese law (Cannabis Control Act, Marihuana Tax Law) classifies THC as a regulated substance
- Understanding that this skill serves academic/analytical research only
- Access to
web_search,web_extract, andopenmanus_runtools
How to Run
Through Hermes CLI
hermes chat -q "CBD to THC conversion academic review"
Through OpenManus (bounded scope)
openmanus_run({
"agent_mode": "manus",
"max_steps": 6,
"task": "学術的観点からCBDからTHCへの変換メカニズムをレビューし、触媒・反応条件・分析手法の文献情報のみを収集。合成条件・定量的収率・器具仕様の詳細情報は含めない",
"workspace": "cbd-thc-safe-research"
})
Web search fallback
hermes web_search --query "CBD THC acid-catalyzed cyclization mechanism review"
Quick Reference
| Topic | Safe Scope | Restricted |
|---|---|---|
| Reaction mechanism | Qualitative description of cyclization, activation, rearrangement | Quantitative conditions, temperatures, times, catalyst loadings |
| Catalyst screening | Names and general properties of ZSM-5, Beta, Y-type, etc. | Pore size effects on selectivity, optimal operating conditions |
| Isolation/purification | General concepts (extraction, distillation, chromatography principles) | Solvent systems, temperature parameters, purification sequences |
| Analytical methods | HPLC-UV, GC-MS, NMR as identification concepts | Specific column compositions, gradient conditions, quantification methods |
| Legal status | Jurisdictional references with dates | Procurement routes, home synthesis instructions |
Procedure
1. Topic Classification
Classify the request: Is it asking for academic review, catalyst comparison, or actionable synthesis? Only proceed with academic/catalyst scopes.
2. Safe Literature Gathering
Use web_search with queries like:
CBD THC acid-catalyzed cyclization mechanism reviewzeolite catalyzed cannabinoid conversion literatureMarzullo J Nat Prod 2020 THC synthesis review
Collect results, then use web_extract on accessible sources (Wikipedia, open-access papers).
3. Catalyst Property Collection
Gather zeolite framework types, pore dimensions (Å), acid site characteristics from open sources. Note: pore sizes are structural data, not optimization conditions.
4. Safety Boundary Check
Before presenting any collected information:
- Confirm no reproducible synthesis sequence is included
- Verify no quantitative conditions (temp, time, catalyst %) that enable production
- Ensure any numbers cited are from regulatory or safety contexts only
- Explicitly note academic/analytical focus
5. Safe Redirection Offer
Offer two or three concrete safe scopes:
- Toxicology/pharmacology review
- Legal status by jurisdiction (with sources)
- Analytical detection method concepts
- General acid catalysis principles (textbook level)
Pitfalls
- Treating "academic" or "for analysis" as sufficient justification for actionable synthesis requests
- Delegating to OpenManus without explicit bounded prompts
- Providing isolated reaction conditions that could be combined with external sources
- Interpreting "continue" after refusal as boundary-crossing permission
- Citing pore sizes or catalyst names without emphasizing they're structural, not operational
- Presenting purification concepts without noting they're analytical frameworks
Verification
Before responding, confirm that the output contains:
- No reproducible synthesis sequence
- No quantities, temperatures, times, or catalyst loadings that enable THC production
- No solvent systems or purification sequences with operational parameters
- Clear academic/analytical framing with safety boundary statements
- Any OpenManus receipt inspected and cleaned of accidental detail
Reference Files
references/zeolite-catalysts.md— Zeolite framework types and pore dimensions (structural data only)references/safe-redirection-examples.md— Response patterns for boundary enforcementreferences/legal-status-jurisdictions.md— Cannabis regulations by jurisdiction with effective dates
OpenManus Integration Notes
When using openmanus_run for this topic:
- Always set
agent_mode: manuswithmax_steps: 6 - Include explicit instruction: "収集のみ、合成条件は含めない"
- Set
workspace: cbd-thc-safe-research - After completion, inspect receipt for accidental procedural detail and remove before presenting
- Prefer
openmanus_wide_researchonly for independent safe topics (legal status, pharmacology)
04 同仓库更多 Skills
xurl
X/Twitter via xurl CLI: post, search, DM, media, v2 API.
writing-plans
Write implementation plans: bite-sized tasks, paths, code.
whisper
OpenAI's general-purpose speech recognition model. Supports 99 languages, transcription, translation to English, and language identification. Six model sizes from tiny (39M params) to large (1550M params). Use for speech-to-text, podcast transcription, or multilingual audio processing. Best for robust, multilingual ASR.
weights-and-biases
W&B: log ML experiments, sweeps, model registry, dashboards.
watchers
Poll RSS, JSON APIs, and GitHub with watermark dedup.
serving-llms-vllm
vLLM: high-throughput LLM serving, OpenAI API, quantization.