codanna

A code-exploration add-on that helps an agent understand a codebase by searching for concepts, symbols, and relationships between functions.

In plain words
What is it for?
Use it to trace how features work, find where concepts or functions are implemented, discover callers and dependencies, and assess the impact of changing code.
Why use it?
It gives code analysis a structured starting point instead of relying only on basic text searches and file reading.

Cursor rule for Cursor

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 rules/cafreeman/foundry-mcp/codanna
Clone the repo
git clone --depth 1 https://github.com/cafreeman/foundry-mcp

Made for: Cursor.

Per session 1,445 This file is loaded in full into every session.
When invoked 1,445 The same file — it is already loaded in full.
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 $0.01445 $0.01445
Opus 5 $0.00723 $0.00723
Sonnet 5 $0.00289 $0.00289
Haiku 4.5 $0.00145 $0.00145

Measured 2d ago against content hash 0aee8377fb91, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

codanna 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.

.cursor/rules/codanna.mdc · 170 lines

How it starts

The opening of the file, as written. The whole thing — 170 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Codanna MCP Tools - Primary Code Exploration Method

CRITICAL: Use Codanna Tools for ALL Code Analysis

When any request involves understanding, exploring, or analyzing code:

  1. ALWAYS start with Codanna tools before using grep/read_file
  2. Use semantic search first - it's more powerful than text-based search
  3. Chain tools together to build complete understanding
  4. Only fall back to basic tools if Codanna doesn't provide results

Required Tool Usage Patterns

🔍 EXPLORATION TRIGGERS

Use Codanna tools when user asks:

  • "How does [feature] work?"
  • "Where is [concept] implemented?"
  • "What calls/uses [function]?"
  • "Find all [pattern] in the code"
  • "Analyze impact of changing [symbol]"
  • "Understand [architectural concept]"

🚀 STANDARD WORKFLOW

Always follow this sequence:

  1. Start Investigation: mcp_codanna_get_index_info (understand scope)
  2. Find Concepts: mcp_codanna_semantic_search_docs (natural language search)
  3. Locate Symbols: mcp_codanna_find_symbol or mcp_codanna_search_symbols (exact/fuzzy search)
  4. Map Relationships: mcp_codanna_get_calls + mcp_codanna_find_callers (dependencies)
  5. Assess Impact: mcp_codanna_analyze_impact (change planning)

🎯 TOOL SELECTION LOGIC

For conceptual questionsmcp_codanna_semantic_search_docs

  • "authentication flow", "error handling patterns", "validation logic"

For specific symbolsmcp_codanna_find_symbol

  • Known function/struct/trait names

For fuzzy explorationmcp_codanna_search_symbols

  • Partial names, exploring patterns, filtering by type

For understanding behaviormcp_codanna_get_calls

  • "What does this function do internally?"

For impact analysismcp_codanna_find_callers + mcp_codanna_analyze_impact

  • "What would break if I change this?"

MANDATORY Usage Rules

✅ DO THIS:

  • Start EVERY code exploration with mcp_codanna_get_index_info
  • Use semantic search with descriptive queries: "HTTP request parsing with validation"
  • Chain tools: semantic_search → find_symbol → get_calls → find_callers
  • Use lower thresholds (0.3-0.4) for broader semantic search results
  • Explain your tool choices to the user

Read the full file on GitHub · 170 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. 2d ago First seen · 170 lines · 1,445 tokens per session scan A 0aee8377fb91

Subscribe to this mod's changes

codanna is a cursor rule published in the GitHub repository cafreeman/foundry-mcp (4 stars, last pushed 5mo ago), licensed MIT. It adds 1,445 tokens to every session, about $0.0072 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.