Skills

Praxist Takeover Codex

Onboard, initialize or repair, validate, and launch a Praxist research task in...

Install

npx skills add sapientinc/PRAXIST --skill praxist-takeover-codex -g -a claude-code -y
Repository
sapientinc/PRAXIST5.5k
Category
Writing
Updated
1mo ago
License
See repository
Context cost
Mediumabout 3,000 tokens
Runs scripts
No, instructions only

What it does

Onboard, initialize or repair, validate, and launch a Praxist research task in Codex-native mode through the official Codex SDK runtime and the operator's existing saved ChatGPT login, with catalog-verified gpt-5.6-luna as the default model and without requesting, storing, or using an API key. Use when a user wants a low-interaction no-key Praxist takeover, explicitly requests Codex-native mode, has no provider key, or invokes this skill with no additional text after already logging in to Codex.

Praxist Takeover: Codex-native mode

Run the complete Praxist takeover workflow while making Codex-native mode the explicit operator choice. The mode uses the operator's saved ChatGPT login. The user invokes this skill once; the current agent performs the checks and shell operations. Do not ask the user to type Praxist commands, provider names, model names, or API keys.

Composition Contract

Load the currently installed praxist-takeover SKILL.md and follow its complete current workflow. In particular:

  • run complete onboarding first;
  • use the complete current task-initialization or repair workflow when needed;
  • preserve evaluator, protocol, lane-routing, resource, and launch-readiness checks;
  • preserve the user's explicit protocol-intent permissions, including an intentionally reduced or incomplete authoritative mode; Codex-native mode changes runtime/authentication only and must not strengthen task semantics;
  • preserve the task-initialization closing-policy regression and positive mature normal-close gate whenever the task defines close-grade evidence;
  • validate with praxist resolve;
  • launch detached, confirm status, and report the independent monitor command.
  • preserve the common agent.reasoning_effort policy; use max unless the user explicitly selected another supported value.

Do not copy or reconstruct an abbreviated takeover from memory. This skill overrides only authentication/runtime selection and the unambiguous-path confirmation behavior described below. If a base takeover instruction conflicts with this file on those points, this file wins.

Treat invocation of praxist-takeover-codex as the user's explicit choice of:

Read the whole SKILL.md on GitHub

More in Writing