Skip to content

Set up AI Hero engineering skills - #3

Draft
keeganmoody33 wants to merge 1 commit into
mainfrom
agent/setup-ai-hero-skills
Draft

Set up AI Hero engineering skills#3
keeganmoody33 wants to merge 1 commit into
mainfrom
agent/setup-ai-hero-skills

Conversation

@keeganmoody33

@keeganmoody33 keeganmoody33 commented Jul 23, 2026

Copy link
Copy Markdown
Owner

What changed

  • Installed Matt Pocock's 22 promoted engineering and productivity skills for Codex
  • Added repository-level agent guidance
  • Configured GitHub Issues, default triage labels, and the existing single-context domain documentation layout
  • Added the generated skills lockfile

Why

This gives PROPER-RESPECT a repeatable, repository-aware workflow for product grilling, domain modeling, specifications, ticketing, TDD, implementation, debugging, and review.

Validation

  • Confirmed all 22 skills are discoverable as project-scoped Codex skills
  • Ran git diff --check
  • Verified the published commit contains all 71 setup files

Greptile Summary

Adds a repository-scoped AI engineering workflow.

  • Installs 22 engineering and productivity skills with OpenAI metadata and a generated lockfile.
  • Adds repository guidance for domain documentation, GitHub issue tracking, and triage labels.
  • Introduces reusable templates and process documentation for planning, implementation, debugging, testing, and review.

Confidence Score: 3/5

The PR should not merge until the installed skills use delegation instructions supported by the target Codex runtime.

Several promoted workflows depend on mandatory Agent calls and named subagent types that the added OpenAI configuration does not provide, preventing those workflows from completing when invoked.

.agents/skills/code-review/SKILL.md, .agents/skills/improve-codebase-architecture/SKILL.md, and .agents/skills/codebase-design/DESIGN-IT-TWICE.md

Important Files Changed

Filename Overview
.agents/skills/code-review/SKILL.md Defines a two-axis review flow, but its mandatory delegation step uses agent interfaces not configured for the target Codex environment.
.agents/skills/improve-codebase-architecture/SKILL.md Adds an architecture exploration workflow whose initial exploration similarly depends on an unconfigured agent type.
docs/agents/issue-tracker.md Configures GitHub issue operations, ticket relationships, fallbacks, and the repository target consistently.
docs/agents/domain.md Documents the existing root glossary and ADR layout and instructs agents how to consume its vocabulary.
skills-lock.json Records the source paths and hashes for all 22 installed skills.

Flowchart

%%{init: {'theme': 'neutral'}}%%
flowchart LR
  Idea[Idea or request] --> Grill[Grill with docs]
  Grill --> Spec[Specification]
  Spec --> Tickets[GitHub tickets]
  Tickets --> Implement[Implementation with TDD]
  Implement --> Review[Code review]
  Bugs[Reported bugs] --> Triage[Triage]
  Triage --> Tickets
  Fog[Large uncertain effort] --> Wayfinder[Wayfinder decisions]
  Wayfinder --> Spec
Loading

Fix All in Devin Fix All in Codex Fix All in Claude Code

Prompt To Fix All With AI
Fix the following 1 code review issue. Work through them one at a time, proposing concise fixes.

---

### Issue 1 of 1
.agents/skills/code-review/SKILL.md:66
**Unsupported delegation interface**

When this mandatory review step runs, it requests an `Agent` tool with `general-purpose` subagents even though the added Codex configuration exposes no such interface, causing the workflow to stop before either review is produced. The same broken delegation mechanism is used by the architecture exploration and design-it-twice flows.

Reviews (1): Last reviewed commit: "Set up AI Hero engineering skills" | Re-trigger Greptile

Greptile also left 1 inline comment on this PR.


- The full diff command and commit list.
- The list of standards-source files you found in step 3, **plus the smell baseline from step 3** pasted in full — the sub-agent has no other access to it.
- The brief: "Report — per file/hunk where relevant — (a) every place the diff violates a documented standard: cite the standard (file + the rule); and (b) any baseline smell you spot: name it and quote the hunk. Distinguish hard violations from judgement calls — documented-standard breaches can be hard, but baseline smells are always judgement calls, and a documented repo standard overrides the baseline. Skip anything tooling enforces. Under 400 words."

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

P1 Unsupported delegation interface

When this mandatory review step runs, it requests an Agent tool with general-purpose subagents even though the added Codex configuration exposes no such interface, causing the workflow to stop before either review is produced. The same broken delegation mechanism is used by the architecture exploration and design-it-twice flows.

Prompt To Fix With AI
This is a comment left during a code review.
Path: .agents/skills/code-review/SKILL.md
Line: 66

Comment:
**Unsupported delegation interface**

When this mandatory review step runs, it requests an `Agent` tool with `general-purpose` subagents even though the added Codex configuration exposes no such interface, causing the workflow to stop before either review is produced. The same broken delegation mechanism is used by the architecture exploration and design-it-twice flows.

How can I resolve this? If you propose a fix, please make it concise.

Fix in Devin Fix in Codex Fix in Claude Code

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant