research

A command that captures research findings from the current conversation and saves them as project knowledge.

In plain words
What is it for?
It helps record findings from research, debugging, and experiments, optionally focused on a particular topic.
Why use it?
It keeps investigation results and debugging discoveries from being lost in chat history.

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/vladolaru/claude-code-plugins/research
Clone the repo
git clone --depth 1 https://github.com/vladolaru/claude-code-plugins
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 795 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 $0.00795
Opus 5 $0.00009 $0.00398
Sonnet 5 $0.00004 $0.00159
Haiku 4.5 $0.00002 $0.00080

Measured 2d ago against content hash ee5a9a5b729c, 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.

plugins/dex/commands/research.md · 78 lines

How it starts

The opening of the file, as written. The whole thing — 78 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/dex:research

Capture research findings from the current conversation. Self-contained — extracts from chat history, no arguments needed. Optional focus hint narrows what to extract.

Step 1: Discover Project Infrastructure

Follow the Project Discovery steps from the knowledge-capture skill.

If .claude/docs/ does not exist, use AskUserQuestion:

Question: "No knowledge directory found. Create it?" Options:

  • Yes, create .claude/docs/ — scaffolds learnings/, patterns/, decisions/, research/
  • Not now — abort capture

If "Not now", stop here. If "Yes", create directories with mkdir -p and continue.

Step 2: Extract Research from Conversation

Run the <pre_extraction_analysis> from the knowledge-capture skill on the relevant conversation exchange. If $ARGUMENTS contains a focus hint, narrow extraction to that topic.

If the conversation contains nothing extractable as research (no investigation, no empirical findings), say so briefly and stop. Do not fabricate knowledge.

Following the Knowledge Extraction from Conversation guidance in the knowledge-capture skill:

  1. Identify the topic — what was being investigated or debugged?
  2. Draft a title as a short descriptive statement (e.g., "PHP 8.3 readonly property behavior with WooCommerce hooks")
  3. Draft a Summary — 2-3 sentence overview of key findings
  4. Identify the Environment — specific versions and configs (e.g., "PHP 8.3.4, WooCommerce 9.6.0") that determine whether findings still apply
  5. Draft What Works — proven approaches with evidence (commands, configs, or code that succeeded)
  6. Draft What Doesn't Work — failed approaches and WHY they failed
  7. Draft Key Findings — detailed empirical observations, numbered for reference
  8. Draft Reproduction Steps — how to verify or reproduce the findings
  9. Draft Open Questions — unresolved issues or areas needing further investigation
  10. Identify 3-5 tags from the technical domain
  11. Determine the filename: YYYY-MM-DD-slug.md

Read the full file on GitHub · 78 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 · 78 lines · 18 tokens per session scan A ee5a9a5b729c

Subscribe to this mod's changes

research is a command published in the GitHub repository vladolaru/claude-code-plugins (8 stars, last pushed 4d ago), licensed MIT. It adds 18 tokens to every session and 795 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.