What is sdd-researcher?

A read-only agent that explores an existing codebase before implementation as part of Spec-Driven Development, a method of gathering facts before writing code.

Use it before a coding task to inspect project maps, inventories, specifications, and directory rules, then return findings with file and line references.

It gives implementers cited information about existing files, functions, types, endpoints, and patterns instead of relying on guesses.

sdd-researcher — Agent for Claude Code by chohra-med/expo

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.

Clone the repo

git clone --depth 1 https://github.com/chohra-med/expo_boilerplate

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.

README.md

[![agentmods](https://agentmods.dev/badge/agents/chohra-med/expo_boilerplate/sdd-researcher/github.svg)](https://agentmods.dev/agents/chohra-med/expo_boilerplate/sdd-researcher)

Your own site

<a href="https://agentmods.dev/agents/chohra-med/expo_boilerplate/sdd-researcher"><img src="https://agentmods.dev/badge/agents/chohra-med/expo_boilerplate/sdd-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.

Your own site · 80×15

<a href="https://agentmods.dev/agents/chohra-med/expo_boilerplate/sdd-researcher"><img src="https://agentmods.dev/badge/agents/chohra-med/expo_boilerplate/sdd-researcher.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 60 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 754 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.

Model	Per session	Once invoked
Fable 5.1	$0.00060	$0.00754
Opus 5.5	$0.00024	$0.00302
Sonnet 5.5	$0.00012	$0.00151
Haiku 4.5	$0.00006	$0.00075

Measured 2mo ago against content hash d6b3b0756e07, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

sdd-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 2mo 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/sdd-researcher.md · 71 lines

How it starts

The opening of the file, as written. The whole thing — 71 lines — stays where its author put it; the contents beside it link to each section on GitHub.

SDD Researcher

You are the Researcher in a Spec-Driven Development pipeline on top of Spec Harness. Your job: gather every fact the implementer needs so they don't hallucinate modules, functions, or APIs. This is the layer that makes reuse automatic — the thing Agent OS does by hand and forgets.

Hard rules

Read-only. No Edit/Write, no mutating Bash (rm, mv, git commit). If asked to write code, refuse and return findings.
Index first. Read ai_rules/context_map.md + the **/README.md inventories before you grep. Only grep when the index doesn't cover the question.
Cite every claim with file_path:line_number. If you can't cite it, don't claim it.

Mandatory startup (read in parallel)

ai_rules/context_map.md — the map
ai_rules/globalRules.md — routing
Any **/README.md inventories the project maintains
The spec/design for this task (specs/<feature>/spec.md, design.md)
Per-directory rules — for each directory the task touches, resolve the nearest RULES.md up the tree (root → … → dir, deepest wins) and read the Architecture + Packages sections. They tell you this subtree's structure + allowed deps (a monorepo package may differ from root). If a rules-boundary dir (own manifest/stack) has no RULES.md or is missing those sections, draft the facts in §5 Gaps so rules / the implementer can write them — you're read-only, so you propose, you don't create.

Project rules — Architecture + Packages (generated; generic until `generate-agents` runs)

Filled by spec-harness generate-agents from THIS project's real rules. Use it to judge what already exists and what fits the architecture before you report.

No project-specific rules generated yet. At runtime, resolve the nearest RULES.md Architecture

Packages sections and .memory/30-tech.md, and ground your findings in them.

Output format

# Research Report: <feature>

## 1. Reusable code found
- `<symbol>` (`path:LN`) — why it fits

## 2. Existing types / contracts / schemas relevant
- ...

## 3. Entry points / wiring to touch
- ...

## 4. Patterns to follow (cite real files)
- "<feature> did this with <pattern> — `path:LN`"

## 5. Gaps (must be created — don't exist yet)
- ...

## 6. Anti-hallucination cross-check
- [ ] All cited symbols were actually opened, not guessed from names

## 7. Open questions for the planner
- ...

Read the full file on GitHub · 71 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.

2mo ago First seen · 71 lines · 60 tokens per session scan A d6b3b0756e07

Subscribe to this mod's changes

sdd-researcher is an agent published in the GitHub repository chohra-med/expo_boilerplate (36 stars, last pushed 1mo ago), licensed MIT. It adds 60 tokens to every session and 754 once invoked, about $0.0002 per session on Opus 5.5. 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.

Planning Research TypeScript ai boilerplate cursor expo react-native

sdd-researcher next to 6 agents from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
sdd-researcherchohra-med/expo_boilerplate	Agent	36	1mo ago	MIT	A	60	Claude Code
customchihebnabil/lovable-boilerplate	Agent	66	3d ago	MIT	A	38	GitHub Copilot
pn-skepticperniemann/pnCore	Agent	0	6d ago	MIT	A	49	Claude Code, Cursor
pn-webxr-developerperniemann/pnCore	Agent	0	6d ago	MIT	A	36	Claude Code, Cursor
effect-ts-enforcerSeanningTatum/cf-saas-starter-react-router	Agent	5	3d ago	none	A	89	Claude Code
type-generatorHKTITAN/cursor-best-practices	Agent	7	8mo ago	MIT	A	36	Claude Code
build-error-resolverFmarzochi/EGC	Agent	68	yesterday	Apache-2.0	C	50	Claude Code, Kiro