reddit-pain-research-skill

KO | EN

gitlite — search

#editor #large-language-models #postgresql #ai-agent #markdown #golang #kotlin #web #command-line #java

reddit-pain-research-skill

(★ 12)

Codex skill for evidence-based customer pain point and monetization plan discovery from Reddit discussions

File Explorer

ZIP Download

- - openai.yaml
- - schemas.md
- - build_reports.py
  - create_research_config.py
- SKILL.md
- test_helpers.py
.gitignore
LICENSE
README.md

# Use via CDN

jsDelivr

jsDelivr serves any public GitHub repository as a CDN with zero setup. Pick a version and a file to get a ready-to-paste link and snippet.

Version

File

Link

Example

Command Glossary

Commands referenced in this DOCs, explained below.

🔍

`python3`

View Details ▼

This command is an alias of `python`.

tldr python

View documentation for the original command:

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

A manually invoked Codex skill for evidence-based customer discovery using
Reddit discussions. It plans a study, collects and clusters pain-point evidence,
and produces falsifiable monetization hypotheses without Firecrawl, a local
model, or project-specific dependencies.

## Requirements

- Codex with skill and web-research support
- Python 3.9 or newer
- Internet access for Reddit and web research

## Install

Copy the installable skill directory into your Codex skills directory:

```sh
cp -R reddit-pain-research "${CODEX_HOME:-$HOME/.codex}/skills/"
```

Restart or open a new Codex session if the skill does not appear immediately.
The skill is manual-only and will not trigger implicitly.

## Use

Invoke it explicitly:

```text
Use $reddit-pain-research to research recurring invoicing pain among freelance
designers in 2026.
```

Optional controls can be expressed naturally:

- Publication year
- Maximum results examined per search query
- Maximum pain-point clusters
- Output directory
- Resume from an existing `research_config.json` or `analysis.json`

For a new run, Codex creates the output directory or asks once whether to
overwrite known research artifacts when the chosen slug already exists. It
then presents a prominent action block with a clickable `research_config.json`
link. The user can inspect or edit the plan and must explicitly approve it
before evidence collection begins.

## Outputs

The workflow keeps one artifact per stage:

- `research_config.json` — reviewed research plan
- `analysis.json` — sourced, clustered, and scored pain evidence
- `monetization_report.md` — ranked hypotheses, scorecards, contrary evidence,
  and falsification tests

## Architecture

```mermaid
flowchart TD
    U[User invokes the skill] --> O[Resolve run options]
    O --> R{Resume from an artifact?}

R -- No --> P[Prepare output directory]
    P --> W[Research communities and plan searches]
    W --> C[create_research_config.py]
    C --> RC[(research_config.json)]
    RC --> G{User approves plan?}
    G -- Revise --> C
    G -- Approved --> E[Collect Reddit evidence with web research]

R -- research_config.json --> E
    E --> A[Deduplicate, analyze, and cluster pain points]
    A --> AJ[(analysis.json)]
    AJ --> S[build_reports.py validates and scores analysis]

R -- analysis.json --> M[Form monetization hypotheses]
    S --> M
    M --> MR[(monetization_report.md)]

SR[references/schemas.md] -. schemas and scoring rules .-> C
    SR -. schemas and scoring rules .-> A
    SR -. schemas and scoring rules .-> M
```

## Safety and evidence rules

- Existing directories require one directory-level overwrite decision.
- Overwrite removes only the three artifacts owned by the skill.
- JSON updates use atomic replacement to preserve the previous file if writing
  is interrupted.
- Reddit sources are deduplicated and must use traceable URLs.
- Pain does not imply willingness to pay; commercial evidence is scored
  separately.
- Conclusions are hypotheses, never claims of business validation.

## Validate

Run the dependency-free regression suite:

```sh
python3 -m unittest discover -s tests -v
```

Validate the skill structure with Codex's `skill-creator`
`quick_validate.py` helper when it is available.

## Repository layout

```text
.
├── .gitignore
├── README.md
├── LICENSE
├── reddit-pain-research/
│   ├── SKILL.md
│   ├── agents/openai.yaml
│   ├── references/schemas.md
│   └── scripts/
│       ├── build_reports.py
│       └── create_research_config.py
└── tests/test_helpers.py
```

## Need something similar?

I build custom AI agents, skills for AI agent harnesses such as OpenClaw and
Codex, n8n and AI-assisted automations, production applications, APIs,
integrations, and scalable software. I also help improve performance and
modernize legacy systems. If you need something similar, please
[get in touch](https://haseebeqx.com/contact-me/).

reddit-pain-research-skill

Owner Repos

File Explorer

# Use via CDN

Command Glossary

`python3`

Was this content helpful?

Recent Feedback

reddit-pain-research-skill

Owner Repos

File Explorer

# Use via CDN

Command Glossary

python3

Was this content helpful?

Recent Feedback

Download README

`python3`