Skip to content
Back to search

agentic-dev-workflow

MarketplaceWorks with Claude CodeGitHub

Two Claude Code skills for agentic development, with an eval harness

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/fuggerstadt1907--agentic-dev-workflow/badge.svg)](/listings/fuggerstadt1907--agentic-dev-workflow)

HTML

<a href="/listings/fuggerstadt1907--agentic-dev-workflow"><img src="/api/listings/fuggerstadt1907--agentic-dev-workflow/badge.svg" alt="findsafeskills safety rating" /></a>

Install

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

Inside Claude Code:

/plugin marketplace add fuggerstadt1907/agentic-dev-workflow
/plugin install agentic-dev-workflow@agentic-dev-workflow

Other ways (copying the skills, one project only) and updating: [install guide](docs/install.md).

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

.claude-plugin/marketplace.json · Manifest · 17 lines
  1. {
  2. "name": "agentic-dev-workflow",
  3. "owner": {
  4. "name": "Alessandro Orlandi",
  5. "url": "https://github.com/fuggerstadt1907"
  6. },
  7. "metadata": {
  8. "description": "Marketplace for the agentic-dev-workflow plugin: grill-ticket and implement-ticket."
  9. },
  10. "plugins": [
  11. {
  12. "name": "agentic-dev-workflow",
  13. "source": "./",
  14. "description": "grill-ticket and implement-ticket: from ticket to requirements to reviewed, committed stages."
  15. }
  16. ]
  17. }
README.md · README · 100 lines
  1. # agentic-dev-workflow
  2. Two Claude Code skills that take a ticket to working, reviewed code. `grill-ticket` checks the
  3. facts itself, asks you only for decisions, and writes a requirements document with stages.
  4. `implement-ticket` builds one stage at a time, has a fresh subagent review it, and commits.
  5. > Evals in progress – results pending.
  6. ```mermaid
  7. flowchart LR
  8. B[Brief / ticket] --> G["/grill-ticket"] --> R[requirements.md]
  9. R --> P
  10. subgraph I["/implement-ticket, once per stage"]
  11. P[Plan] --> S[Thin slice] --> V[Fresh-context review]
  12. V --> F[Fix] --> C[Commit]
  13. end
  14. C -. next stage .-> P
  15. C --> W[Worklog]
  16. ```
  17. Brief → /grill-ticket → requirements → /implement-ticket per stage (plan → thin slice →
  18. fresh-context review → fix → commit, repeat) → worklog
  19. ## Install
  20. Inside Claude Code:
  21. ```text
  22. /plugin marketplace add fuggerstadt1907/agentic-dev-workflow
  23. /plugin install agentic-dev-workflow@agentic-dev-workflow
  24. ```
  25. Other ways (copying the skills, one project only) and updating: [install guide](docs/install.md).
  26. ## Usage
  27. ```text
  28. /grill-ticket #123 → docs/…/requirements.md (Status: ready)
  29. /implement-ticket docs/…/requirements.md → stage 1 built, reviewed, committed
  30. /clear
  31. /implement-ticket docs/…/requirements.md → stage 2 …
  32. ```
  33. The short form (`/implement-ticket`) works as long as no other installed skill has the same name.
  34. The namespaced form always works: `/agentic-dev-workflow:grill-ticket`,
  35. `/agentic-dev-workflow:implement-ticket`.
  36. The skills are written in English and work in the language of the project they run in. Each step
  37. is explained on the [workflow page](docs/workflow.md). Templates for a project's
  38. [`CLAUDE.md`](templates/CLAUDE.md) and [`WORKLOG.md`](templates/WORKLOG.md) are in `templates/`.
  39. The goal is that you understand the workflow in 2 minutes and install it in 5. Nobody has measured
  40. that yet.
  41. ## Prevention and detection
  42. Some rules in `implement-ticket` prevent defects: a security-relevant stage gets acceptance
  43. criteria written from an attacker's view before building, and every test that asserts something
  44. is absent gets a positive check that what it looks for is really in the input. The review eval does not measure
  45. these rules.
  46. Detection is the fresh-context review with
  47. [`REVIEW.md`](skills/implement-ticket/REVIEW.md). The review eval measures it against a
  48. plain-prompt baseline, using planted defects and clean diffs.
  49. ## Why this exists
  50. Coding agents write code fast. They also build on unclear requirements and fill the gaps with
  51. assumptions; the agent that wrote the code reviews it and keeps its blind spots; and nobody
  52. measures whether a prompt or skill helps. These skills come out of a real project. The evals, still
  53. in progress, compare each skill with a plain prompt and report where the skill does not win, too.
  54. This repo is built with its own skills. Every stage is recorded in the
  55. [worklog](docs/WORKLOG.md), and the decisions are in the [requirements](docs/requirements.md).
  56. ## Credits
  57. These ideas come from Matt Pocock's [skills](https://github.com/mattpocock/skills):
  58. - [`grilling`](https://github.com/mattpocock/skills/blob/abeeaf95660ba9e8b7c753eb38f441ca707a8823/skills/productivity/grilling/SKILL.md):
  59. the design tree worked in rounds (`grill-ticket` calls it the decision tree), the frontier of questions whose prerequisites are settled, and
  60. the split between facts the agent looks up and decisions the user makes. Used in `grill-ticket`
  61. and, as the split between decisions the agent may settle and decisions for the user, in
  62. `implement-ticket`.
  63. - [`tdd`](https://github.com/mattpocock/skills/blob/abeeaf95660ba9e8b7c753eb38f441ca707a8823/skills/engineering/tdd/SKILL.md):
  64. tests at agreed seams, each with a one-line note on what it catches and what it misses, and red
  65. before green. Used in `implement-ticket`'s plan and build steps.
  66. - [`to-tickets`](https://github.com/mattpocock/skills/blob/abeeaf95660ba9e8b7c753eb38f441ca707a8823/skills/engineering/to-tickets/SKILL.md):
  67. tracer bullets, narrow slices that cut through every layer. Used in `implement-ticket`'s build
  68. step.
  69. No text is copied from these skills.
  70. ## Contributing
  71. See [CONTRIBUTING.md](CONTRIBUTING.md).
  72. ## Licence
  73. [MIT](LICENSE)

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.