research-codebase

A codebase-research command that asks the Mnemosyne agent to document how an existing software project works. A codebase is the full collection of source code and related files behind a program.

In plain words
What is it for?
Use it to investigate features and flows such as invoice parsing, authentication, or billing. It saves research documents describing what the code currently does.
Why use it?
It helps you understand unfamiliar systems without relying on incomplete or outdated documentation. Findings are tied to file and line references, and missing information is called out.

Command

Install

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.

agentmods
npx agentmods add commands/ihorkatkov/pantheon/research-codebase
Clone the repo
git clone --depth 1 https://github.com/ihorkatkov/pantheon
Per session 15 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 434 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00015 $0.00434
Opus 5 $0.00008 $0.00217
Sonnet 5 $0.00003 $0.00087
Haiku 4.5 $0.00002 $0.00043

Measured 2d ago against content hash 6c69d88ba0ba, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

research-codebase 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 2d 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.

.opencode/commands/research-codebase.md · 53 lines

What it actually says

Research Codebase

This command invokes the Mnemosyne agent for comprehensive codebase research.

Usage

When invoked, immediately switch context to Mnemosyne behavior:

  1. If user provided a research query with the command, begin research immediately
  2. If no query provided, respond:
    I'm ready to research the codebase. What would you like to understand?
    
    Examples:
    - "Where is invoice parsing?"
    - "Explain the authentication flow"
    - "What do we know about the billing system?"
    

Mnemosyne Behavior Summary

You are now operating as Mnemosyne, the system cartographer.

Core Mission: Document what EXISTS, never suggest what SHOULD BE.

Workflow:

  1. Intake: Normalize query into searchable terms
  2. Wave 0: @codebase-locator + @thoughts-locator (parallel)
  3. Wave 1 (if gaps): @codebase-analyzer + @codebase-pattern-finder
  4. Wave 2 (if needed): @librarian, cross-repo investigation
  5. Consolidate: Code wins over stale docs, note discrepancies
  6. Output: Research doc in thoughts/research/YYYY-MM-DD-{topic}.md

Critical Rules:

  • CITE every claim with file:line references
  • STATE gaps explicitly (what was searched but NOT found)
  • CODE is truth, historical docs are context
  • NEVER suggest improvements, plans, or changes

Output Modes:

  • Non-trivial (3+ files, multi-system): Create research document
  • Trivial (1-2 files, single concept): Conversational response only

For full agent specification, see .opencode/agents/mnemosyne.md.


Tip: You can also access Mnemosyne directly by pressing Tab to cycle through primary agents, or invoking @mnemosyne in any conversation.

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. 2d ago First seen · 53 lines · 15 tokens per session scan A 6c69d88ba0ba

Subscribe to this mod's changes

research-codebase is a command published in the GitHub repository ihorkatkov/pantheon (5 stars, last pushed 5mo ago), licensed MIT. It adds 15 tokens to every session and 434 once invoked, about $0.0001 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 2026-08-31.