What is praxis-drive?

A workflow for repeatedly working through an active task list without waiting for a user after every step. It can run iterations in the current session or through an unattended runner.

Use it to continue a multi-step implementation, resume paused work, update the task ledger, and stop after a configured task, slice, phase, or review gate.

It reduces the need to manually restart the coding agent for each small delivery step. It also checks configured limits, stopping conditions, and progress records while work continues.

praxis-drive — Skill for Claude Code and Codex by jeet129

Install

Getting it into your agent

It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.

Clone the repo

git clone --depth 1 https://github.com/jeet129/praxis

agentmods

npx agentmods add skills/jeet129/praxis/praxis-drive

Made for: Claude Code, Codex.

Or install praxis, the plugin that ships this one along with the rest of its 105 skills, 12 commands, 17 agents, 6 hooks.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/skills/jeet129/praxis/praxis-drive/github.svg)](https://agentmods.dev/skills/jeet129/praxis/praxis-drive)

Your own site

<a href="https://agentmods.dev/skills/jeet129/praxis/praxis-drive"><img src="https://agentmods.dev/badge/skills/jeet129/praxis/praxis-drive/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/jeet129/praxis/praxis-drive"><img src="https://agentmods.dev/badge/skills/jeet129/praxis/praxis-drive.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 57 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 994 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits

NVIDIA SkillSpector pass 7 Sept 2026

How audits are shown

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.

Model	Per session	Once invoked
Fable 5.1	$0.00057	$0.00994
Opus 5	$0.00028	$0.00497
Sonnet 5	$0.00011	$0.00199
Haiku 4.5	$0.00006	$0.00099

Measured 10d ago against content hash 0dc9f1f8255d, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

praxis-drive 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 10d 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.

codex-plugin-assets/skills/praxis-drive/SKILL.md · 50 lines

How it starts

The opening of the file, as written. The whole thing — 50 lines — stays where its author put it; the contents beside it link to each section on GitHub.

capability: command
domain: codex-plugin
state: experimental
dependencies: [agentic-harness-orchestration, adaptive-model-routing]
triggers: [praxis drive, keep going autonomously, run this without me, resume drive, autonomous drive]
outputs: [updated_task_ledger, slice_close_summary, stop_flags, model_routing_log]
consumers: [praxis-slice, praxis-review]
references: [../../governance/autonomy.yaml, ../../references/loop-contracts.md]

Praxis Drive

Start or resume autonomous drive. There are two paths and they loop differently: the unattended runner loops for you (each codex exec it fires is one fresh delivery-lead iteration, and the runner re-invokes until a stop); the in-session path is a loop YOU run — you keep issuing iterations within this session until a stop boundary. Do not conflate the two: "one iteration then exit, no internal loop" applies to a delivery-lead invoked BY the runner, NOT to the in-session orchestrator, which must continue.

Steps

Check the dial. Read ../../governance/autonomy.yaml. Note stop_after (task | slice | phase | gate), run_budget, and stall.max_iterations_without_ledger_change. This is the OPTIONAL human boundary — three non-negotiable stops (decision points, governance gates, budget/stall/exhaustion) fire regardless of this setting.
Locate the active ledger. Look for .project/working/slice-<id>-tasks.yaml. If none exists, route through praxis-slice first — delivery-lead → lead-developer produce the packet AND the ledger (per ../../references/loop-contracts.md §2) before drive can start. Never fabricate a ledger.
State the budget and stops once. Before iterating, tell the user (once): the stop_after dial, the run_budget caps, the three non-negotiable stops, and that slice-close summaries post to .project/telemetry/summaries/ regardless of the dial.
Run it. Offer both paths (a long-running / "run without me" request implies unattended):
- Unattended (deterministic loop): point the user at scripts/praxis-drive.sh --project-dir . --harness codex, which invokes the harness headlessly, loops iteration-to-iteration itself, and enforces run_budget caps + stall detection outside this session's context. Each codex exec it fires is exactly one delivery-lead iteration; the runner handles the looping. This is the reliable way to drive a whole slice/workflow.
- In-session (you run the loop): iterate the drive protocol yourself, continuously, within this session. Each pass is delivery-lead (codex-agents/delivery-lead.toml) executing exactly one autonomous-drive iteration — read ledger + named task context, dispatch/complete, run verify, update status/attempts, apply model routing — then report the outcome and immediately begin the next pass. Do NOT stop after one task and wait: keep going through the slice drain (full-ceremony code review, security review, QA, closure evidence) and on into the next slice's ledger. Per skills/autonomous-drive "stop_after semantics", per-slice review/QA are part of the drain, NOT governance gates — under stop_after: gate they do not end the run. Stop cleanly ONLY when a non-negotiable stop (decision point, governance gate, budget/stall/exhaustion) or the dial's stop_after boundary is actually reached.

Read the full file on GitHub · 50 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.

10d ago First seen · 50 lines · 57 tokens per session scan A 0dc9f1f8255d

Subscribe to this mod's changes

praxis-drive is a skill published in the GitHub repository jeet129/praxis (7 stars, last pushed 6d ago), licensed MIT. It adds 57 tokens to every session and 994 once invoked, about $0.0003 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.

Building agents Planning

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

local-ai-agents

Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…

microsoft/ai-agents-for-beginners · 200 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

insight-error-page

Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…

vercel/next.js · 83 tokens

next-cache-components-optimizer

Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…

vercel/next.js · 170 tokens

next-partial-prefetching-adoption

Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…

vercel/next.js · 103 tokens