{
  "markdown": "<p align=\"center\">\n  <a href=\"https://dynos.fit\">\n    <img src=\"dynos-logo-dark.svg\" alt=\"dynos.fit\" />\n  </a>\n</p>\n\n# dynos-work\n\nA Claude Code and Codex plugin that builds features, catches its own bugs, and gets better at your codebase over time.\n\n## Install\n\n### Claude Code\n\n```\nclaude plugin marketplace add dynos-fit/dynos-work\nclaude plugin install dynos-work\n```\n\n### Codex\n\nThis repo also includes a Codex plugin manifest at `.codex-plugin/plugin.json`. Install it through your Codex plugin workflow from this repository root.\n\n## Use\n\n### Claude Code\n\n```\n/dynos-work:start build a user settings page\n```\n\n### Codex\n\nInvoke the `start` skill with your task, for example: \"Use dynos-work start to build a user settings page.\"\n\nThat's it. You'll answer a few questions about what you want, approve a spec, approve a plan — then it builds.\n\n## What happens\n\nWhen you start a task, the plugin:\n\n1. **Asks what you need** — a few targeted questions to nail down scope\n2. **Writes a spec** — exact behavior, edge cases, what's out of scope. You approve it.\n3. **Writes a plan** — which files change, in what order, how it'll be tested. You approve it.\n4. **Builds in parallel** — multiple agents work on independent pieces simultaneously\n5. **Audits its own work** — independent agents check for bugs, security issues, and spec gaps\n6. **Repairs what it finds** — automatically, up to 3 attempts before it asks you\n\nAfter the task, it learns from what went wrong and builds prevention rules for next time. Your 10th task is better than your 1st.\n\n## Commands\n\n```\n/dynos-work:start [task]       start a new task\n/dynos-work:execute            run the approved plan\n/dynos-work:audit              audit, repair, and close\n/dynos-work:status             check where your task is\n/dynos-work:resume             continue after interruption\n/dynos-work:investigate [bug]  deep bug investigation\n/dynos-work:residual           inspect or drain non-blocking findings\n```\n\n## Requirements\n\nClaude Code or Codex, plus Python 3 for the local dynos-work control plane.\n\n## Running with permissions ON\n\nThe pipeline works without `--dangerously-skip-permissions`. Every deterministic step funnels through one binary — `<plugin-root>/bin/dynos` — so you can allow that single command prefix once (via the permission prompt's \"don't ask again\" or a `Bash(<plugin-root>/bin/dynos ...)` allow rule in your settings) and then you only interact with the decisions that are actually yours: discovery questions, spec approval, plan approval, and TDD review. Plugin guardrail denials are not Claude Code permission errors — they say so explicitly and name the sanctioned alternative. Details: `docs/write-boundary-spec.md`.\n\n## Links\n\n- [Changelog](CHANGELOG.md)\n- [Internals & CLI reference](INTERNALS.md)\n- [Architecture](ARCHITECTURE.md)\n- [Pipelines](PIPELINES.md)\n\n---\n\n> Completion is a control decision, not a model opinion.\n",
  "bytes": 2916,
  "sha": "7723108a778b5e3d9205818a9ba226472d145909a3cd8e02dc74c0aa7755d10b",
  "repo_slug": "hassamsheikh/dynos-audit",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/plg_hassamsheikh_dynos_audit_dynos_audit_6e5f2a4d/readme"
}