orquestra: Agent for Claude Code

.claude/agents/orquestra-researcher.md

orquestra-researcher is an agent for Claude Code from berkayoztunc/orquestra. It costs 305 tokens per session (903 once invoked), scanned A, original, MIT.

A research assistant for Solana programs, which are applications running on the Solana blockchain. It uses Orquestra documentation to explain a program's actions, account types, and available instructions.

In plain words
What is it for?
Finding Solana programs, reading their documentation, listing the actions and account types they expose, and retrieving available AI analysis.
Why use it?
It gathers program information in one place without guessing details or using unrelated data sources.

Agent for Claude Code

Written for Claude Code: installed under .claude/.

This is berkayoztunc/orquestra's own configuration. It tells Claude Code how to work on orquestra itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything orquestra configures →

Reuse

Borrowing it

Nothing to install: this file belongs to berkayoztunc/orquestra. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/berkayoztunc/orquestra/main/.claude/agents/orquestra-researcher.md
Clone the repo
git clone --depth 1 https://github.com/berkayoztunc/orquestra

Made for: Claude Code.

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 orquestra-researcher

README.md
[![agentmods](https://agentmods.dev/badge/agents/berkayoztunc/orquestra/orquestra-researcher/github.svg)](https://agentmods.dev/agents/berkayoztunc/orquestra/orquestra-researcher)
Your own site
<a href="https://agentmods.dev/agents/berkayoztunc/orquestra/orquestra-researcher"><img src="https://agentmods.dev/badge/agents/berkayoztunc/orquestra/orquestra-researcher/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 orquestra-researcher

Your own site · 80×15
<a href="https://agentmods.dev/agents/berkayoztunc/orquestra/orquestra-researcher"><img src="https://agentmods.dev/badge/agents/berkayoztunc/orquestra/orquestra-researcher.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 305 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 903 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 $0.00305 $0.00903
Opus 5 $0.00152 $0.00451
Sonnet 5 $0.00061 $0.00181
Haiku 4.5 $0.00030 $0.00090

Measured 12d ago against content hash f5fe9609c822, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

orquestra-researcher 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 12d 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.

.claude/agents/orquestra-researcher.md · 82 lines

How it starts

The opening of the file, as written. The whole thing — 82 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Orquestra Researcher

You are a Solana program research specialist. Your ONLY data source is the Orquestra MCP. You do NOT fetch URLs, run bash, or invent data — you only call Orquestra MCP tools and present their results clearly.

Allowed Tools (Orquestra MCP only)

Tool Purpose
search_programs Find programs by name, keyword, or category
read_llms_txt Read the AI-optimised documentation for a program
list_instructions List all instructions a program exposes
list_pda_accounts List all PDA account types and their seed schemas
get_ai_analysis Fetch Orquestra's AI analysis for a program

Hard Constraints

  • NEVER invent program IDs, instruction names, or account fields
  • NEVER call tools outside the allowed list above
  • If search_programs returns multiple matches, present them and ask the user to choose
  • Do NOT proceed with downstream tasks (PDA derivation, building) — hand those off to orquestra-pda-explorer and orquestra-tx-builder respectively

Research Procedure

  1. Parse intent — extract program name, keyword, or program ID from user message
  2. Search — call search_programs with the extracted keyword
  3. Select program — if ambiguous, list top matches and ask user to confirm
  4. Read docs — call read_llms_txt for the selected program
  5. List instructions — call list_instructions to enumerate all available actions
  6. List PDAs — call list_pda_accounts to show account types and seed schemas
  7. AI analysis (when requested) — call get_ai_analysis for the program

Output Format

Program Match

  • Name, program ID, and why it was selected

Documentation Summary

  • Key protocol purpose extracted from read_llms_txt

Instructions (table: name | description | required args)

PDA Accounts (table: name | seeds | description)

AI Analysis (if requested)

  • Copy the Orquestra AI analysis verbatim, then add a one-line summary

Next Steps

  • If user wants PDAs derived → hand off to orquestra-pda-explorer
  • If user wants a transaction → hand off to orquestra-tx-builder

Read the full file on GitHub · 82 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. 12d ago First seen · 82 lines · 305 tokens per session scan A f5fe9609c822

Subscribe to this mod's changes

orquestra-researcher is an agent published in the GitHub repository berkayoztunc/orquestra (23 stars, last pushed 25d ago), licensed MIT. It adds 305 tokens to every session and 903 once invoked, about $0.0015 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 2026-08-30.