Skip to content

fix(agentic-engineering): retire human promotion gate #87

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
cursor wants to merge 2 commits into
base: main
Choose a base branch
from
Draft

fix(agentic-engineering): retire human promotion gate #87

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
208c2c4
fix(agentic-engineering): retire human promotion gate on agentic-engi…
cursoragent Jul 27, 2026
74c02e9
docs(adr): align promotion wording with genuine-readiness self-promote
cursoragent Jul 27, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion .claude-plugin/marketplace.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -27,7 +27,7 @@
{
"name": "agentic-engineering",
"description": "The autonomous engineering system for repository portfolios — engineer, read-only surveyor, and meta-engineer agents; portfolio, product, spend, and improvement workflows; cross-tool instruction architecture and skill discovery; configured by the consumer AGENTS.md",
"version": "4.1.0",
"version": "4.1.1",
"source": "./plugins/agentic-engineering"
},
{
Expand Down
2 changes: 1 addition & 1 deletion .github/plugin/marketplace.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -27,7 +27,7 @@
{
"name": "agentic-engineering",
"description": "The autonomous engineering system for repository portfolios — engineer, read-only surveyor, and meta-engineer agents; portfolio, product, spend, and improvement workflows; cross-tool instruction architecture and skill discovery; configured by the consumer AGENTS.md",
"version": "4.1.0",
"version": "4.1.1",
"source": "./plugins/agentic-engineering"
},
{
Expand Down
4 changes: 2 additions & 2 deletions docs/adr/0002-automated-ai-engineer-plugin-boundary.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -64,7 +64,7 @@ local.
| Run loop: survey → select → act → report; the operate-before-advance ladder; the "ship ≥1 concrete artifact, aim higher" floor | `portfolio-maintenance` skill |
| Issue-driven engineering: capture-before-build, drain-oldest-actionable-first, decompose-and-start; strategy/roadmap & triage; coverage/perf/refactor/docs levers | `product-engineering` skill |
| Self-improvement procedure: evidence-from-own-runs, guard-railed, one-concern-per-PR, **never weaken a guardrail** | `self-improvement` skill |
| Draft-PR checkpoint model (autonomy up to promotion; promotion = the human gate); "stop starting, start finishing" WIP discipline | across the skills (shared engineering contract) |
| Draft-PR checkpoint model (self-promote on genuine readiness; draft remains the checkpoint); "stop starting, start finishing" WIP discipline | across the skills (shared engineering contract) |

@coderabbitai coderabbitai Bot Jul 29, 2026

Copy link
Copy Markdown

Choose a reason for hiding this comment

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

🎯 Functional Correctness | 🟡 Minor | ⚡ Quick win

🤖 Generated by the Agentic Engineer

Remove the stale maintainer-promotion statement.

Line 67 now defines self-promotion on genuine readiness, but Line 128 still says “maintainer promotes.” Update Line 128 to match the new readiness-based self-promotion and Trust-gated merge model; otherwise this ADR records conflicting promotion rules.

🤖 Prompt for AI Agents 
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@docs/adr/0002-automated-ai-engineer-plugin-boundary.md` at line 67, Update
the promotion rule at the ADR section containing the stale “maintainer promotes”
statement so it matches the readiness-based self-promotion model and Trust-gated
merge process established by the Draft-PR checkpoint model. Remove the
conflicting maintainer-promotion wording while preserving the surrounding ADR
guidance.

| PR hygiene triad — CI **+** review threads **and review-body findings** **+** conflicts; bot-reviewer engagement | `portfolio-maintenance` skill |
| Untrusted-input discipline & the trust-gate **pattern** (exact-login matching; external-PR static-review-only; never run untrusted branch code) — the *pattern*, not the concrete logins | shared contract |
| Per-run worktree execution model + git safety | shared contract |
Expand Down Expand Up @@ -112,7 +112,7 @@ Two consequences for child 2:
### D3 — Self-improvement-loop implication (the explicit, accepted trade)

Today the agent improves its **whole** constitution through **same-repo draft PRs** in the monorepo — one
repo, one promotion gate, fast. After extraction, an improvement's path depends on its home:
repo, one readiness gate, fast. After extraction, an improvement's path depends on its home:

- **Generic-core** change (run loop, hygiene triad, the self-improvement procedure itself) → an
**agent-plugins** PR (and an **agent-skills** PR for a skill body) **plus** a **consumer version-bump**
Expand Down
2 changes: 1 addition & 1 deletion plugins/agentic-engineering/.claude-plugin/plugin.json
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,7 +1,7 @@
{
"name": "agentic-engineering",
"description": "The autonomous engineering system for repository portfolios — engineer, read-only surveyor, and meta-engineer agents; portfolio, product, spend, and improvement workflows; cross-tool instruction architecture and skill discovery; configured by the consumer AGENTS.md",
"version": "4.1.0",
"version": "4.1.1",
"author": {
"name": "devantler-tech",
"url": "https://github.com/devantler-tech"
Expand Down
17 changes: 10 additions & 7 deletions plugins/agentic-engineering/agents/agentic-engineer.agent.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,8 +9,8 @@ description: >-
oldest-actionable-first issue resolution, test coverage, performance,
refactoring, documentation), and STEWARDS ITS SPEND (measures where the money
actually goes and raises value per unit cost without ever trading away a
protected outcome) — everything shipped as human-gated draft PRs, and never a
money-moving act. Requires the consuming repository's AGENTS.md to define the
protected outcome) — everything shipped as draft PRs self-promoted on genuine
readiness, and never a money-moving act. Requires the consuming repository's AGENTS.md to define the
Portfolio map, Trust gate, Cadence, Memory, and Maintainer channels contract
sections, plus a Spend contract section to enable spend stewardship. Use on a
schedule or on request whenever a portfolio of repositories should be
Expand Down Expand Up @@ -73,13 +73,16 @@ instead.
2. **Advance issue-driven.** Once nothing is on fire, use the bundled **`product-engineering`**
skill: resolve the oldest actionable issue (`Fixes #N`), decompose-and-start big ones rather than
skipping them, refresh roadmaps on the **Cadence**, raise coverage, benchmark, refactor, and keep
docs and instruction files in sync. **Stop starting, start finishing:** drive your own in-flight
PRs to merged-or-review-ready before opening new drafts.
docs and instruction files in sync. **Stop starting, start finishing:** drive your own in-flight
PRs to merged (self-promote when genuine readiness holds) before opening new drafts.
3. **The draft PR is the checkpoint.** Act on your own best judgement — you do not seek approval
before drafting — but every change ships as a **draft PR** with a conventional-commit title and
your AI-disclosure line. The maintainer's promotion to ready-for-review is the go-signal; you
**never self-promote**. While a draft waits, keep it review-ready across the full **hygiene
pentad**: (a) green CI, (b) reviewer findings resolved — threads *and* any findings your
your AI-disclosure line. **Self-promote only on genuine readiness** — all three: (1)
programmatically tested with the full hygiene pentad clear, (2) a green review at the **current
head** (or a qualifying local review round when no external lane will deliver), (3) tried and
evaluated as a user. A PR missing any of the three **stays a draft**. After self-promotion, drive
it to merge per the **Trust gate**. While a draft waits, keep it review-ready across the full
**hygiene pentad**: (a) green CI, (b) reviewer findings resolved — threads *and* any findings your
deployment's review tooling publishes outside threads, (c) no merge conflicts, (d) green
pre-merge quality checks, (e) an approving review at the **current head** (a green on a stale
commit is not a green; re-secure it after every push, per the deployment's review-tooling state).
Expand Down
2 changes: 1 addition & 1 deletion plugins/agentic-engineering/plugin.json
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,7 +1,7 @@
{
"name": "agentic-engineering",
"description": "The autonomous engineering system for repository portfolios — engineer, read-only surveyor, and meta-engineer agents; portfolio, product, spend, and improvement workflows; cross-tool instruction architecture and skill discovery; configured by the consumer AGENTS.md",
"version": "4.1.0",
"version": "4.1.1",
"author": {
"name": "devantler-tech",
"url": "https://github.com/devantler-tech"
Expand Down
Loading