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.
npx agentmods add instructions/butttons/dora/agents-mdgit clone --depth 1 https://github.com/butttons/doraWhat 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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00833 | $0.00833 |
| Opus 5 | $0.00417 | $0.00417 |
| Sonnet 5 | $0.00167 | $0.00167 |
| Haiku 4.5 | $0.00083 | $0.00083 |
Grade A, and why
dora AGENTS.md 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.
How it starts
The opening of the file, as written. The whole thing — 81 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
dora is a CLI that converts SCIP indexes into a queryable SQLite database. AI agents use it to answer questions about large codebases without reading files or tracing imports manually.
Stack
- Runtime: Bun
- Database: SQLite via
bun:sqlite - Language: TypeScript
- Protobuf parsing:
@bufbuild/protobuf - AST parsing:
web-tree-sitter(on-demand, per file) - Output format: TOON by default (
dora status), JSON via--json(dora status --json)
Source layout
src/
├── commands/ # one file per CLI command
├── converter/ # SCIP protobuf parser + SQLite converter
├── db/ # schema and all SQL queries
├── mcp/ # MCP server, tool definitions, handlers
├── schemas/ # Zod schemas and inferred types
├── tree-sitter/ # grammar discovery, parser, language registry
└── utils/ # config, errors, output formatting
Two indexing layers
SCIP — runs the configured indexer (e.g. scip-typescript), produces a .scip protobuf, converts it to SQLite. Gives you symbols, references, and file-to-file dependencies derived from actual import resolution.
Tree-sitter — parses source files on-demand using wasm grammars. Covers what SCIP doesn't: function signatures, cyclomatic complexity, class hierarchy, code smells. Grammar discovery checks local node_modules, then global bun packages, then explicit config paths.
Database design
Denormalized counts (symbol_count, dependency_count, dependent_count, reference_count) are pre-computed at index time. Most queries are index lookups, not aggregations.
Local symbols (function parameters, closure variables) are flagged is_local = 1 and filtered out by default. Symbol kinds are extracted from SCIP documentation strings since scip-typescript doesn't populate the kind field.
Schema: src/converter/schema.sql. All queries: src/db/queries.ts.
Config file: .dora/config.json
{
"root": "/absolute/path/to/repo",
"scip": ".dora/index.scip",
"db": ".dora/dora.db",
"commands": {
"index": "scip-typescript index --output .dora/index.scip"
},
"lastIndexed": "2025-01-15T10:30:00Z",
"ignore": ["test/**", "**/*.generated.ts"],
"treeSitter": {
"grammars": {
"typescript": "/explicit/path/to/tree-sitter-typescript.wasm"
}
}
}
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.
- 2d ago First seen · 81 lines · 833 tokens per session scan A ec296c574c67
dora AGENTS.md is an instructions file published in the GitHub repository butttons/dora (108 stars, last pushed 5mo ago), licensed MIT. It adds 833 tokens to every session, about $0.0042 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.
Other instructions, from other repositories
roam-code AGENTS.md
Instructions for Cranot/roam-code, covering agents.md — roam-code development guide, what this project is, documentation hub, where files go (private vs public) and quality discipline (from internal/dogfood/ + agi-in-md).
codeseek CLAUDE.md
Instructions for CodeBendKit/codeseek, covering claude.md, build & test, enter rust project directory, build and run tests.
bonsai-ninja AGENTS.md
Instructions for gromhacks/bonsai-ninja, covering bonsai-ninja, map a codebase, optional explicit semantic sidecar prewarm, explicit spelling for default syntax/construct indexing and optional during active editing.
ScalaSemantic AGENTS.md
Instructions for MercurieVV/ScalaSemantic, covering agents.md instructions and quick start.
ScalaSemantic CLAUDE.md
Instructions for MercurieVV/ScalaSemantic, covering scalasemantic, stack, layout, architecture and mcp tools (target surface).
infigraph CLAUDE.md
Claude Code instructions for intuit/infigraph, covering claude.md, working convention, what this is, build, test, lint and architecture.