How do I install the Richardgill Nix Oracle skill?

Run npx skillmds@latest add tomevault-io/richardgill-nix-oracle in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

What does the Richardgill Nix Oracle skill do?

Oracle It is listed under Coding & Dev Tools on SkillMD.

Is Richardgill Nix Oracle safe to use?

This skill has not completed SkillMD's automated safety review yet. Independent scanners report: SkillSpector: CAUTION, Skill Scanner: PASS. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Richardgill Nix Oracle?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Richardgill Nix Oracle free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Richardgill Nix Oracle?

tomevault-io (@tomevault-io) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Richardgill Nix Oracle by tomevault-io

Name: Richardgill Nix Oracle
Author: tomevault-io

File contents

Oracle

Escalate to GPT-5.4 with high reasoning effort for problems requiring deeper analysis. Use for:

Complex debugging with elusive bugs
Architectural decisions and tradeoffs
Security audits and vulnerability analysis
Code review requiring deep reasoning

Main command

Run Codex with GPT-5.4 in non-interactive mode with high reasoning:

codex exec -m gpt-5.4 --sandbox read-only -c model_reasoning_effort='"high"' "$PROMPT"

Where $PROMPT is your analysis request. Codex bundles relevant context from your repo automatically.

Reasoning effort

high — deep reasoning, ~3x tokens, best for complex problems (oracle default)
medium — balanced (codex default)
low — fast, minimal thinking

Options

--sandbox read-only — analyze without modifying files (recommended)
--sandbox workspace-write — allow file modifications if needed
-o /tmp/oracle-response.md — save response to file
-C <dir> — run in specific directory

Guidelines

Oracle is read-only — analyzes but doesn't modify code
Treat outputs as advisory — verify against codebase
Use for genuinely complex problems (not routine tasks)
High reasoning uses ~3x tokens but thinks deeper

Examples

# Debug intermittent auth issue
codex exec -m gpt-5.4 --sandbox read-only -c model_reasoning_effort='"high"' \
  "The auth flow fails intermittently. Check src/auth/ for race conditions."

# Architectural review
codex exec -m gpt-5.4 --sandbox read-only -c model_reasoning_effort='"high"' \
  "Review src/api/ data flow and suggest improvements for scalability."

# Security audit
codex exec -m gpt-5.4 --sandbox read-only -c model_reasoning_effort='"high"' \
  "Audit src/handlers/ for OWASP top 10 vulnerabilities."

What to ask oracle:

$ARGUMENTS

Converted and distributed by TomeVault — claim your Tome and manage your conversions.

---name: richardgill-nix-oracledescription: Oracle---# OracleEscalate to GPT-5.4 with high reasoning effort for problems requiring deeper analysis. Use for:- Complex debugging with elusive bugs- Architectural decisions and tradeoffs- Security audits and vulnerability analysis- Code review requiring deep reasoning## Main commandRun Codex with GPT-5.4 in non-interactive mode with high reasoning:```bashcodex exec -m gpt-5.4 --sandbox read-only -c model_reasoning_effort='"high"' "$PROMPT"```Where `$PROMPT` is your analysis request. Codex bundles relevant context from your repo automatically.## Reasoning effort- `high` — deep reasoning, ~3x tokens, best for complex problems (oracle default)- `medium` — balanced (codex default)- `low` — fast, minimal thinking## Options- `--sandbox read-only` — analyze without modifying files (recommended)- `--sandbox workspace-write` — allow file modifications if needed- `-o /tmp/oracle-response.md` — save response to file- `-C <dir>` — run in specific directory## Guidelines- Oracle is **read-only** — analyzes but doesn't modify code- Treat outputs as advisory — verify against codebase- Use for genuinely complex problems (not routine tasks)- High reasoning uses ~3x tokens but thinks deeper## Examples```bash# Debug intermittent auth issuecodex exec -m gpt-5.4 --sandbox read-only -c model_reasoning_effort='"high"' \ "The auth flow fails intermittently. Check src/auth/ for race conditions."# Architectural reviewcodex exec -m gpt-5.4 --sandbox read-only -c model_reasoning_effort='"high"' \ "Review src/api/ data flow and suggest improvements for scalability."# Security auditcodex exec -m gpt-5.4 --sandbox read-only -c model_reasoning_effort='"high"' \ "Audit src/handlers/ for OWASP top 10 vulnerabilities."```## What to ask oracle:$ARGUMENTS---> Converted and distributed by [TomeVault](https://tomevault.io/claim/richardgill) — claim your Tome and manage your conversions.

Frequently asked questions

Richardgill Nix Oracle