research

research is a command for coding agents from jasonhnd/life_OS. It costs 23 tokens per session (597 once invoked), scanned A, original, Apache-2.0.

A command for multi-agent research on a topic. It gathers views from several research roles and combines them into a draft wiki entry, including checks for confirmation bias.

In plain words
What is it for?
Use it for broad or contested topics that need academic, practical, historical, alternative, and critical perspectives before writing a wiki draft.
Why use it?
It reduces the chance that a topic is researched from only one angle or that supporting evidence is accepted without challenge.

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/jasonhnd/life_os/research
Clone the repo
git clone --depth 1 https://github.com/jasonhnd/life_OS

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.

agentmods badge for research

README.md
[![agentmods](https://agentmods.dev/badge/commands/jasonhnd/life_os/research.svg)](https://agentmods.dev/commands/jasonhnd/life_os/research)
Your own site
<a href="https://agentmods.dev/commands/jasonhnd/life_os/research"><img src="https://agentmods.dev/badge/commands/jasonhnd/life_os/research.svg" alt="Measured on agentmods" height="20"></a>
Per session 23 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 597 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 $0.00023 $0.00597
Opus 5 $0.00012 $0.00298
Sonnet 5 $0.00005 $0.00119
Haiku 4.5 $0.00002 $0.00060

Measured 4d ago against content hash 1afcc61ca524, 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 4d 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.

docs/history/v1.10-command-surfaces/scripts/commands/research.md · 61 lines

How it starts

The opening of the file, as written. The whole thing — 61 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/research · Multi-agent research -> wiki draft (v1.8.1 F4)

User invoked: /research $ARGUMENTS

What this does

Spawns parallel general-purpose subagents to research a topic from multiple angles, synthesizes results into a single wiki/<domain>/<slug>.md draft (per wiki/SCHEMA.md), and presents it for user accept/edit/reject.

Default: 5 agents (academic / practitioner / contrarian / origin / adjacent). With --depth deep: 8 agents (adds mechanistic / data-statistics / meta-review). With --thesis "<hypothesis>": thesis-driven mode; agents weight evidence for and against the stated hypothesis.

Total wall time target: <= 7 minutes (5 min parallel agents + 2 min synthesis).

Backup mode

Slash command is backup mode. Primary path: user says "研究 X" / "research X" / "深挖 X"; ROUTER matches those phrases inline and reads scripts/prompts/research.md. Slash command exists for explicit arguments (--depth, --thesis).

Execution

Read and follow scripts/prompts/research.md step-by-step. That prompt defines the 5-phase workflow (decompose -> parallel -> synthesize -> counter-bias -> user-review -> write). Do not invent your own flow.

Argument parsing:

  • First positional non-flag = <topic> (required; reject if missing)
  • --depth deep = use 8 agents instead of 5
  • --thesis "<...>" = thesis-driven mode; treat thesis as the working hypothesis to test

After execution, the user MUST see:

  1. The 5-or-8 agent summaries (<= 500 words each)
  2. The synthesized wiki draft with full frontmatter
  3. The Counterpoints section (always, even if "none found, all sources converge")
  4. Explicit accept / edit / reject prompt
  5. On accept: wiki/log.md updated, file written, wiki/INDEX.md patched

Cost note

This command is expensive (5-8 parallel WebSearch+WebFetch agents). Use it only when:

  • The topic genuinely needs multi-perspective coverage
  • You're seeding a wiki entry from scratch (vs updating)
  • A single-source draft would carry too much confirmation bias

Read the full file on GitHub · 61 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. 4d ago First seen · 61 lines · 23 tokens per session scan A 1afcc61ca524

Subscribe to this mod's changes

research is a command published in the GitHub repository jasonhnd/life_OS (5 stars, last pushed 27d ago), licensed Apache-2.0. It adds 23 tokens to every session and 597 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.