Arno_1

TOOLS.md

Mar 19th, 2026
23
0
Never
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 2.78 KB | None | 0 0
  1. # TOOLS.md - Local Notes
  2.  
  3. Skills define _how_ tools work. This file is for _your_ specifics — the stuff that's unique to your setup.
  4.  
  5. ## Research Tools
  6.  
  7. Your primary tools for research:
  8. - `web_search` — find sources
  9. - `web_fetch` — retrieve and verify sources
  10. - `sessions_spawn` — spawn sub-agents for heavy research
  11.  
  12. ### Key Parameters
  13.  
  14. **web_fetch:**
  15. - Max characters per fetch: 50,000 (system hard limit)
  16. - Workaround for long articles: fetch in sections using anchors or fetch multiple related pages
  17. - Always verify the fetched content matches what you expected
  18.  
  19. **web_search:**
  20. - No hard limit on number of queries
  21. - Use varied phrasing to combat selection bias
  22. - Search for "X criticism" or "X controversy" to find dissenting views
  23. - Include recency in queries: "X 2025" to find latest developments
  24.  
  25. **sessions_spawn (sub-agents):**
  26. - Use for heavy research tasks requiring many searches/fetches
  27. - Configure with `--thinking high` for complex synthesis
  28. - Can use cheaper model via `--model` to control costs
  29. - Default max depth: 1 (you can spawn sub-agents but they cannot spawn further unless configured)
  30. - Sub-agents inherit your tool permissions (but not session tools)
  31.  
  32. ## What Goes Here
  33.  
  34. Things like:
  35. - Current research project notes
  36. - Preferred search query patterns
  37. - Any custom verification steps
  38. - Notes on sources that have been particularly reliable or problematic
  39.  
  40. ## Examples
  41.  
  42. ```markdown
  43. ### Research Notes (Current Projects)
  44.  
  45. **Project: AI Safety Standards 2025**
  46. - Started: 2025-03-17
  47. - Status: In progress
  48. - Key queries: ["AI safety framework 2025", "AI governance latest", "AI ethics criticism"]
  49. - Sources to prioritize: peer-reviewed papers, official standards bodies (ISO, IEEE), government policy docs
  50. - Gaps so far: Non-English perspectives limited
  51. - Sub-agent spawned: yes (session ID: abc123) — awaiting results
  52.  
  53. ### Source Reliability Notes
  54.  
  55. - **Highly reliable:** .gov, .edu, .org (established orgs), peer-reviewed journals
  56. - **Use with caution:** News outlets (check for original sources), industry blogs (conflict of interest likely)
  57. - **Verify carefully:** Social media claims, Reddit posts, anonymous "expert" quotes
  58. - **Avoid:** Unverified PDFs from unknown domains, content farms
  59.  
  60. ### Query Bank
  61.  
  62. For bias mitigation, keep a list of alternative framings:
  63. - Primary: "quantum computing advantages"
  64. - Alternative 1: "quantum computing limitations"
  65. - Alternative 2: "quantum computing criticism"
  66. - Alternative 3: "quantum computing challenges"
  67. ```
  68.  
  69. ## Why Separate?
  70.  
  71. Skills are shared. Your setup is yours. Keeping them apart means you can update skills without losing your notes, and share skills without leaking your infrastructure.
  72.  
  73. ---
  74.  
  75. Add whatever helps you conduct rigorous research. This is your research notebook.
  76.  
  77.  
Advertisement
Add Comment
Please, Sign In to add comment