[developerz-ai]

This pull request establishes the base an AI coding agent needs to work well in developerz-ai/mini.

Stack

The account's model ran this repository and answered its stack: Bash + ShellCheck.

Established

• bin/setup — 4. scripts that work
• bin/dev — 4. scripts that work
• bin/check — 4. scripts that work
• bin/stop — 4. scripts that work
• .claude/agents/model-selection.md — 2. a specialist roster
• .claude/agents/doctor-checks.md — 2. a specialist roster
• .claude/agents/ollama-server.md — 2. a specialist roster
• .claude/agents/opencode-config.md — 2. a specialist roster
• .claude/skills/verify-change/SKILL.md — 3. project skills
• .claude/skills/model-registry/SKILL.md — 3. project skills
• .claude/skills/platform-module/SKILL.md — 3. project skills
• .claude/skills/new-tool/SKILL.md — 3. project skills
• .env.example
• .githooks/pre-commit — 4. scripts that work
• .gitignore
• .mcp.json — 5. agent wiring
• .dz/maintainer/reviewer.md — 0. the repo policy
• .dz/onboarding/scorecard.md

Left alone

These already existed and are yours — nothing here was rewritten, reordered or reflowed:

• CLAUDE.md — 1. a brain
• AGENTS.md
• docs/README.md
• .claude/commands — 9. workflow commands
• .github/workflows/ci.yml — 4. scripts that work
• .dz/maintainer/maintainer.yml — 0. the repo policy

Verification

The gate this pull request declares, run on the box that wrote it: bin/setup then bin/check, before these changes (base) and after them (head); then bin/stop, the teardown the platform runs when every session ends; then bin/setup and bin/stop once more on the same box (reuse), the way every later session meets it.

Proof: proved — every command ran green at head.

• ./bin/setup absent at base
• ./bin/check absent at base
• ./bin/setup exited 0 at head (0.4s)

  se by @jdx – installing 6 tools
  mise ⇢ bun@1.4.4    4ms · already installed
  mise ⇢ go@1.23      1ms · already installed
  mise ⇢ node@24      1ms · already installed
  mise ⇢ python@3.14  4ms · already installed
  mise ⇢ ruby@3.4.11  5ms · already installed
  mise ⇢ uv@0.13      1ms · already installed
  mise ████████████████ 6/6 · installed 0 tools · 6 already installed in 42ms
  mise all tools are installed
  

• ./bin/check exited 0 at head (86.3s)

   E: a relative member path resolves against the runner's repo root
  ok   F: every member runs even when an earlier one failed
  ok   G: a member printing FAIL or SKIP but exiting 0 is reported 'PASS'
  ok   H: the summary holds one section and one line per member, and is appended to
  ok   I: with GITHUB_STEP_SUMMARY empty or unset the runner runs and writes no summary
  PASS: run-tests.sh verdict contract
  

• ./bin/stop exited 0 at head (0.1s)

  nothing to stop
  

• ./bin/setup exited 0 at reuse (0.2s)

  ise by @jdx – installing 6 tools
  mise ⇢ bun@1.4.4    3ms · already installed
  mise ⇢ go@1.23      0ms · already installed
  mise ⇢ node@24      0ms · already installed
  mise ⇢ python@3.14  2ms · already installed
  mise ⇢ ruby@3.4.11  0ms · already installed
  mise ⇢ uv@0.13      0ms · already installed
  mise ████████████████ 6/6 · installed 0 tools · 6 already installed in 9ms
  mise all tools are installed
  

• ./bin/stop exited 0 at reuse (0.1s)

  nothing to stop
  

Scorecard

13 of 15 items ok. The whole card, with every piece of evidence, is .dz/onboarding/scorecard.md.

• policy — ok — .dz/maintainer/maintainer.yml already present, left alone
• brain — ok — CLAUDE.md already present, left alone
• roster — ok — .claude/agents/model-selection.md established
• skills — ok — .claude/skills/verify-change/SKILL.md established
• scripts — ok — bin/setup established
• wiring — ok — .mcp.json established
• commands — ok — .claude/commands already present, left alone
• access — gap — landed 0 of 1 asked
• layout — gap — partial house layout
• smoke_test — ok — tests/test_provider.sh in the repository
• env_example — ok — .env.example established
• agents_md — ok — AGENTS.md already present, left alone
• per_workspace_brains — ok — single-package repository: 0 workspaces
• docs_readme — ok — docs/README.md already present, left alone
• gate_proved — ok — ./bin/check exited 0 at head (86.3s)

First tasks

Queued from what this pass found, and none of them runs before this merges:

• Verify the base runs clean
• Refine CLAUDE.md into a tight project brain
• Align bin/check with the existing CI pipeline (proposal)

Merging

This rewrites how agents read your repo, on a repo the platform has not worked before,
so it is yours to merge — not the bot's. Merging it is also what marks the repo
onboarded; until then coding tasks on it are held (issue triage, PR review and
report-only tasks are unaffected).

---

🤖 Opened by developerz.ai, an automated agent, not a human, on developerz-ai/mini.

Raised automatically by a developerz.ai maintainer lane, from an interactive repository setup session.

🤖 developerz.ai, an automated agent, not a human — this PR is one slice of a stacked plan.
Dispatched automatically by a developerz.ai maintainer lane
Raised automatically by the developerz.ai repo onboarding maintainer lane, from setup.
Plan: https://app.developerz.ai/fleet/plans/tsk_0a0ad43b121f9de3db03300d21a3dd28
Part 1/1
BYOK spend: ~$0.08 estimated from published list prices (byok funding evidence).

[developerz-ai]

🟡 Reviewed

1 actionable comment(s) · grounded on your code

_{⚠️ 1 config note(s): test files were not shown to the reviewer — this diff touched no path recognised as a test, so no assertion was available to read the change against}

_{🤖 developerz.ai — automated review, running on your model and your box.}

[developerz-ai]

Review summary — 19 file(s), 7 finding(s).

Critical 0 · Major 1 · Minor 6 · Nit 0

bin/check's new gate is sound, but the scaffolding overclaims: verify-change reserves exit codes 75/79 bin/check never emits, claims CI and the pre-commit hook run it when neither does at head, and .gitignore omits .env though .env.example says to create it and never commit it.

File	Findings
.claude/skills/verify-change/SKILL.md	1 major, 2 minor
.env.example	1 minor
.gitignore	1 minor
.mcp.json	1 minor
bin/check	1 minor

• test files were not shown to the reviewer — this diff touched no path recognised as a test, so no assertion was available to read the change against

🤖 developerz.ai review — automated, running on your model and your box. What is this?

[developerz-ai]

Address review findings on #40

• Picked up
• Surveying the repo
• Working
• Pull request opened — Establish the AI-first base for developerz-ai/mini #40
• Done

Work landed on #40 — watching it toward merge.

11m · 16 tool calls · 10 messages

---

🤖 developerz.ai, an automated agent, not a human — developerz-ai[bot]

[developerz-ai]

A fix round is working on this pull request. Nothing is needed from you right now.

• Red on this head: no check is failing
• What was done: Fix round tsk_ed0f78f84275f57011186c26f0626b51 is running against the failing check(s) and any open review findings.
• What happens next: It pushes to this branch when it is done; CI and the review then run on the new head.

_{Status for head 03916b18d355. This comment is updated in place.}

🤖 Posted by developerz.ai — an automated agent, not a human.

[developerz-ai]

Review summary — 5 file(s), 1 finding(s).

Critical 0 · Major 0 · Minor 1 · Nit 0

Security verdict: no widening — the increment only closes the previously flagged secret-handling gaps (.env now ignored, MCP env names documented in .env.example) and hardens the gate's exit contract; one issue found: a docs claim that CI grades the gate scripts bin/check now covers.

1 non-blocking finding(s) below major (no thread on a re-review — only blocking findings open threads)

• .claude/skills/verify-change/SKILL.md:12 minor · docs The reword says CI runs the same commands, but bin/check now guards five gate scripts (bin/setup, bin/dev, bin/check, bin/stop, .githooks/pre-commit) missing from CI's bash -n/shellcheck lists (per CLAUDE.md), so green CI no longer grades the gate scripts. Qualify the claim here or extend the workf…

• test files were not shown to the reviewer — this diff touched no path recognised as a test, so no assertion was available to read the change against

🤖 developerz.ai review — automated, running on your model and your box. What is this?