TRIP Research logo

TRIP Research

Community
PiLastDigit
TRIP-research

Exploratory research or spike - investigation without production code

Overview

PublisherPiLastDigit
RepositoryTRIP-workflow
Skill nameTRIP-research
Stars
636
Forks
91
Bundled files
Instructions only
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by PiLastDigit on GitHub. Read the source before you install it.

Installation

Install the TRIP Research AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/PiLastDigit/TRIP-workflow/tree/master/skills/TRIP-research
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/PiLastDigit/TRIP-workflow.git /tmp/TRIP-workflow
mkdir -p .claude/skills
cp -r /tmp/TRIP-workflow/skills/TRIP-research .claude/skills/TRIP-research
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable TRIP Research in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads TRIP Research on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so TRIP Research is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Research Mode

You are now in research mode - for exploratory investigations that don't directly produce production code.

Use this for:

  • Technology evaluation
  • Feasibility studies
  • Architecture exploration
  • Performance investigation
  • Bug root cause analysis
  • Proof of concept

Your Task

Research: $ARGUMENTS


Step 0: Read fully @docs/ARCHI.md

Step 1: Define Scope

1.1 Clarify the Question

"What specific question(s) are we trying to answer?"

Document the research question(s) clearly:

  • What do we need to find out?
  • What would a successful outcome look like?
  • What decisions will this research inform?

1.2 Compute Box

"How much thinking effort does this research require?"

Based on the complexity of the question(s), suggest an appropriate thinking level:

LevelWhen to UseResearch Type
quickSimple lookup, straightforward answer"What's the syntax for X?"
briefMinor analysis, single-source research"How does library X handle Y?"
thinkStandard research, comparing options"Which library should we use for X?"
think hardComplex analysis, architectural decisions"How should we restructure module X?"
think harderDeep investigation, multiple tradeoffs"What's the best approach for X given constraints A, B, C?"
ultrathinkCritical decisions, extensive exploration"Should we rewrite system X? What are all implications?"

Use the AskUserQuestion tool to confirm the thinking level:

  • Question: "Based on this research scope, I suggest using [level] thinking. Does that seem appropriate?"
  • Options: "Yes, use [level]" (proceed with suggested level), "Use a different level" (I want to adjust the thinking effort)

Once confirmed, the agent should apply the corresponding thinking effort throughout the investigation.


Step 2: Research Plan

Create a lightweight research plan (not a full TRIP plan):

markdown
# Research: [Topic]

## Question(s)

- [Primary question]
- [Secondary questions if any]

## Approach

1. [First thing to investigate]
2. [Second thing to investigate]
3. [...]

## Success Criteria

- [ ] [What we need to determine]
- [ ] [What we need to determine]

## Compute Level

[quick / brief / think / think hard / think harder / ultrathink]

Step 3: Confirm & Start

Present the research plan summary to the user, then use the AskUserQuestion tool with two questions:

Question 1 (header: "Start"):

  • Question: "Research plan ready — Question: [primary question], Approach: [brief summary], Compute: [level]. Ready to start?"
  • Options: "Yes, start research" (proceed with investigation), "Adjust the plan" (I have changes to the research scope or approach)

Question 2 (header: "Output"):

  • Question: "How do you want the findings delivered?"
  • Options: "Chat only" (respond directly in the conversation), "Write a memo" (create a file in docs/6-memo/)

If "Adjust": Modify the plan based on user feedback, then re-present using AskUserQuestion.

If "Yes": Proceed with investigation. Remember the output preference for Step 5.


Step 4: Investigation

Conduct the research:

For Technology Evaluation

  • Review documentation
  • Check community/ecosystem health
  • Look at alternatives
  • Consider maintenance burden
  • Assess learning curve

For Feasibility Study

  • Identify constraints
  • Prototype critical parts (throwaway code OK)
  • Identify risks and unknowns
  • Estimate effort

For Performance Investigation

  • Establish baseline metrics
  • Identify bottlenecks
  • Test hypotheses
  • Measure improvements

For Bug Investigation

  • Reproduce the issue
  • Trace the root cause
  • Identify contributing factors
  • Consider similar vulnerabilities

Step 4b: Codex Cross-Check

For decision-grade findings — architecture recommendations, technology choices, anything the user will build on (typically compute level think hard and above) — red-team the draft conclusion with the codex-ask skill before presenting. Skip for quick lookups.

bash
bash .claude/skills/codex-ask/scripts/start.sh \
    --prompt-file .claude/skills/codex-ask/prompts/ask.tpl \
    <topic-label> "Here is my draft recommendation: <summary + key rationale>. Red-team it: what am I missing, what would you choose instead, and why?"

Follow up in the same thread (resume.sh + followup.tpl) if the answer raises points worth probing. Then:

  • Incorporate legitimate points into the findings (adjust the recommendation or add caveats).
  • Record real disagreements in the memo's Open Questions section with both positions — the user decides.
  • This is advisory, not gating: you own the final recommendation.

Step 5: Present Findings

If "Chat only" was selected

Present the findings directly in the conversation, structured as:

  1. Summary (2-3 sentences)
  2. Key Findings (numbered list)
  3. Recommendations (what to do, rationale, alternatives)
  4. Open Questions (if any)

Then use the AskUserQuestion tool:

  • Question: "Research complete. What would you like to do next?"
  • Options: "Elaborate on findings" (dive deeper into specific results), "Save as memo" (write findings to docs/6-memo/ after all), "Plan implementation" (create a TRIP plan based on these results), "Done" (no further action needed)

If "Save as memo": write the memo using the template below, then confirm the file location.

If "Write a memo" was selected

Create findings document in docs/6-memo/:

File: docs/6-memo/research_[date]_[topic].md

markdown
# Research: [Topic]

**Date**: DD-MM-YYYY
**Author**: [Name]

## Summary

[2-3 sentence executive summary]

## Questions Investigated

### Q1: [Question]

**Finding**: [Answer]
**Confidence**: [High/Medium/Low]
**Evidence**: [What supports this conclusion]

### Q2: [Question]

[...]

## Key Findings

1. **[Finding 1]**: [Details]
2. **[Finding 2]**: [Details]
3. **[Finding 3]**: [Details]

## Recommendations

- **Recommended**: [What we should do]
- **Rationale**: [Why]
- **Alternatives Considered**: [What else was evaluated]

## Open Questions

- [Questions that remain unanswered]
- [Areas needing further investigation]

## Next Steps

- [ ] [Action item 1]
- [ ] [Action item 2]

## Appendix (optional)

### Code Snippets / Prototypes

[Any throwaway code created during research]

### References

- [Links to documentation, articles, etc.]

Then use the AskUserQuestion tool:

  • Question: "Research documented at docs/6-memo/research_[date]_[topic].md. What would you like to do next?"
  • Options: "Elaborate on findings" (dive deeper into specific results), "Plan implementation" (create a TRIP plan based on these results), "Done" (no further action needed)

What This Workflow Produces

  • Clear recommendations
  • Basis for future TRIP planning
  • Optionally: documented findings in docs/6-memo/

What This Workflow Does NOT Produce

  • No production code
  • No version bump
  • No changelog entry
  • No commits to main

Transition to Implementation

If research concludes that implementation should proceed, use the AskUserQuestion tool:

  • Question: "Based on this research, would you like me to create a TRIP plan for implementation?"
  • Options: "Yes, create a plan" (use findings to inform TRIP-1-plan), "Not yet" (I'll decide later)

If yes, use findings to inform TRIP-1-plan.

Frequently asked questions

What does the TRIP Research AI skill do?

Exploratory research or spike - investigation without production code

Why use TRIP Research on TypingMind?

Because you install it once and use it with any model. TRIP Research is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install TRIP Research in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/PiLastDigit/TRIP-workflow/tree/master/skills/TRIP-research. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use TRIP Research?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with TRIP Research?

As many as you like. As long as a model supports skills, you can use TRIP Research with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the TRIP Research AI skill free?

It is published on GitHub by PiLastDigit. Check the repository for licensing terms. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇