Instructions file CodexOpenCode
AGENTS.md instructions for bdhrs/vicaya, covering agent guidance — vicaya, working with tests, database edits and static analysis.
Instructions file CodexOpenCode
AGENTS.md instructions for bdhrs/vicaya, covering agent guidance — vicaya, working with tests, database edits and static analysis.
Command
Compare how different translators render a Pāḷi word or phrase, as an easy-to-scan Markdown table.
Command
Write a short plain-English quick-study essay on a topic — the Monarch Notes/CliffsNotes recipe for any subject, no inline citations, saved to the vault.
Command
Process accumulated /vicaya run retrospectives into the improvement backlog, apply channel tuning, and present the top 5 issues to work on.
Command
Audit and repair the Vicaya PDF tree — one PDF twin per note, generate the missing ones, delete the orphans.
Command
Search existing Obsidian vault notes to check if any already partially answer the research question before starting a full /vicaya run.
Command
Deeply research a specific Pāḷi/Buddhist question across local sources plus web and answer in chat with citations — no vault note, no gates.
Command
Full /vicaya research run in the fixed "What the suttas say about X" series format — earliest-strata evidence vs later attributions, one quote per idea.
Command
Run structured Pāḷi/Buddhist research across local sources (Obsidian vault, CST canon, library, web, YouTube) with cross-check and vault note output.
Skill Claude CodeCodex
Compare how different translators render a Pāḷi word or phrase, as an easy-to-scan Markdown table. Use when the user asks to compare/align translations of a Pāḷi term, or types /vicaya-align .
Skill Claude CodeCodex
Write a plain-English essay (a 20–30 minute read) that gives a layperson the main points of a large or complex topic without reading the book/paper/field yourself — a modern Monarch Notes / CliffsNotes pamphlet for any subject. No inline footnotes; ends with a further-reading link list. Saved into the vault. Invoke…
Skill Claude CodeCodex
Process accumulated run retrospectives in runs/ into the prioritized improvement backlog (runs/TODO.md), apply channel tuning, archive processed runs, and present the top 5 issues for the user to pick one to work on. Invoke when the user types /vicaya-improve or asks to "process the runs", "triage run feedback", or…
Skill Claude CodeCodex
Audit and repair the Vicaya PDF tree so every note has exactly one PDF twin and the tree holds nothing else — generates the missing PDFs and deletes the orphans. Invoke when the user types /vicaya-pdf or asks to "check the PDFs", "fix the missing PDFs", "clean up the orphaned PDFs", or "make the PDFs match the notes".
Skill Claude CodeCodex
Search existing Obsidian vault notes to check if any already answer the research question, and recommend /vicaya-quick or /vicaya based on question depth and existing coverage.
Skill Claude CodeCodex
Deeply research a specific Pāḷi/Buddhist question across the user's local sources (Obsidian vault, CST canon, DPD dictionary, library folders, EBC vault, SuttaCentral parallels) plus web/YouTube, and answer directly in chat with citations — no vault note, no gates, no ceremony. Invoke when the user types /vicaya-quick…
Skill Claude CodeCodex
Run the full /vicaya workflow with the fixed "What the suttas say about X" series format — earliest-strata evidence vs later attributions, one quote per idea, written into the series subfolder of the vault. Invoke when the user types /vicaya-what-the-suttas-say or asks for a "what the suttas say about X" note.
Instructions file CodexOpenCode
AGENTS.md instructions for bdhrs/kamma, covering kamma — agent guide, repo layout, working on the prompts and conventions.
Command
Scaffolds the project and sets up the Kamma environment.
Command
Plans a thread and generates thread-specific spec documents.
Command
Executes the tasks defined in the specified thread's plan.
Command
Reviews a thread and gets it ready to finish.
Command
Finalizes a reviewed thread, updates project documentation, and handles cleanup.
Command
Saves session context to the active thread so the next agent can pick up where you left off.
Command
Sweep kamma lessons across every repo, find recurring mistakes, and consolidate them into durable improvements to the kamma framework itself.