research

A command that starts the second phase of an AIDD project: a broad research sweep based on an existing project brief. It records possible choices in an options.json file.

In plain words
What is it for?
Use it after the ideation phase to research the project's technical and product choices, including extra research angles supplied as arguments.
Why use it?
It structures research so decisions are based on multiple searches, trade-offs, and opened sources instead of a small sample of results.

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/rchase999/aidd/research
Clone the repo
git clone --depth 1 https://github.com/rchase999/aidd
Per session 18 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,211 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 $0.00018 $0.01211
Opus 5 $0.00009 $0.00606
Sonnet 5 $0.00004 $0.00242
Haiku 4.5 $0.00002 $0.00121

Measured 2d ago against content hash f145734d0a09, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

research 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.

aidd/commands/research.md · 68 lines

How it starts

The opening of the file, as written. The whole thing — 68 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Phase 2 — Research

Load the research-dossier skill. It is the protocol; this command is the trigger.

Extra angles requested: $ARGUMENTS

Read .aidd/01-ideate/brief.md first. If it is missing, stop and run /aidd:ideate.

!node "${CLAUDE_PLUGIN_ROOT}/scripts/aidd.mjs" phase research 2>&1

The standard

Sweep, don't sample. The user asked for everything findable, so absence of an option in the gate means you could not find it — not that you did not look. Concretely: for each segment, keep searching until two consecutive fresh searches turn up nothing you haven't already seen.

Every option carries pros, cons, and at least one real source URL you actually opened. An option with no cons means you didn't read far enough.

If the run is in brownfield mode

Check first:

!node "${CLAUDE_PLUGIN_ROOT}/scripts/aidd.mjs" scan 2>&1

An existing codebase has already answered most of the questions a greenfield gate would ask. Offering them again is noise that trains the user to click through the gate without reading it — which destroys the only thing this framework is for.

So before deriving segments:

  1. Read .aidd/00-context/codebase.md. If it does not exist, dispatch aidd-surveyor now.
  2. Split every candidate decision into closed (the codebase committed to it) and open (genuinely undecided).
  3. Closed decisions become "locked": true segments with a lockedReason citing the file that proves it. They render read-only, are excluded from the submit requirement, and reach the contract as LOCKED.
  4. Put the stack, constraints, and must-not-break list into the top-level context block of options.json. Every context.stack entry needs evidence — the file that proves it — or validation fails.
  5. Research the open segments as hard as you would greenfield, but filter every option against the locked constraints. An option that needs Python 3.12 in a repo pinned to 3.11 is not an option; it is a distraction. If it is genuinely worth the pin change, present it and price the change honestly.

Read the full file on GitHub · 68 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 · 68 lines · 18 tokens per session scan A f145734d0a09

Subscribe to this mod's changes

research is a command published in the GitHub repository rchase999/aidd (2 stars, last pushed 23d ago), licensed MIT. It adds 18 tokens to every session and 1,211 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.