stage-10-pr

stage-10-pr is a skill for Claude Code from cdeust/ai-architect-mcp. It costs 4 tokens per session (2,269 once invoked), scanned A, original, MIT.

A delivery workflow that prepares a tested code change for a GitHub pull request, including records that trace it back to the original problem.

In plain words
What is it for?
Use it to push a branch, create a merge-ready pull request, attach labels and audit references, and close a tracked finding.
Why use it?
It removes the need to assemble the pull request description and change history by hand. It also blocks delivery when the required test report is missing.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin.

Part of the ai-architect plugin — 13 skills, 4 commands, 2 MCP servers, 1 plugin shipped together

Good fit Use it to push a branch, create a merge-ready pull request, attach labels and audit references, and close a tracked finding.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/cdeust/ai-architect-mcp/stage-10-pr
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add cdeust/ai-architect-mcp --skill stage-10-pr
Clone the repo
git clone --depth 1 https://github.com/cdeust/ai-architect-mcp

Made for: Claude Code.

Or install ai-architect, the plugin that ships this one along with the rest of its 13 skills, 4 commands, 2 MCP servers, 1 plugin.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for stage-10-pr

README.md
[![agentmods](https://agentmods.dev/badge/skills/cdeust/ai-architect-mcp/stage-10-pr/github.svg)](https://agentmods.dev/skills/cdeust/ai-architect-mcp/stage-10-pr)
Your own site
<a href="https://agentmods.dev/skills/cdeust/ai-architect-mcp/stage-10-pr"><img src="https://agentmods.dev/badge/skills/cdeust/ai-architect-mcp/stage-10-pr/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for stage-10-pr

Your own site · 80×15
<a href="https://agentmods.dev/skills/cdeust/ai-architect-mcp/stage-10-pr"><img src="https://agentmods.dev/badge/skills/cdeust/ai-architect-mcp/stage-10-pr.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 4 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,269 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00004 $0.02269
Opus 5 $0.00002 $0.01135
Sonnet 5 $0.00001 $0.00454
Haiku 4.5 $0.00000 $0.00227

Measured 10d ago against content hash fc5e7098ad3c, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

stage-10-pr scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/stage-10-pr/SKILL.md · 238 lines

How it starts

The opening of the file, as written. The whole thing — 238 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Allostatic Priming

You are a delivery agent. Push. Create PR. Write audit trail. Close finding. Your output is a merge-ready pull request with a complete audit trail linking every change back to the original finding, through the PRD, to the verification report. The PR description is machine-generated from stage artifacts — not hand-written.

Trigger

USE WHEN: create PR, pull request, push branch, deliver, merge ready, audit trail, GitHub PR, sha256 manifest, delivery, close finding, PR creation NOT FOR: code implementation — stage 6, verification — stage 7, benchmarking — stage 8, test execution — stage 9

Survival Question

"Is the PR open on GitHub with correct labels, body, and audit trail reference?"

Before you start

  1. ai_architect_load_context(stage_id=9, finding_id="{findingID}") — load test report (confirms all tests passed)
  2. ai_architect_load_session_state(session_id="{sessionID}") — confirm currentStage = 10
  3. Load all stage artifacts for PR body assembly

Missing Stage 9 test report = BLOCK. Cannot create PR without passing tests.

Input contract

Field Type Source Required
stage-9-test-report.json JSON StageContext[stage-9] YES — BLOCK if missing
stage-7-hor-report.json JSON StageContext[stage-7] YES — for PR body
stage-2-impact-map.json JSON StageContext[stage-2] YES — impact score for PR body
prd-overview.md MD StageContext[stage-4] YES — PRD overview for PR body
Implementation branch git ref pipeline/{findingID} YES — branch to push

Operations

1. Push branch to remote

ai_architect_git_push(
  branch="pipeline/{findingID}"
)
→ Push implementation branch to GitHub
→ Failure = BLOCK — network issue, auth expired, or branch conflict

2. Create pull request

Assemble PR body from stage artifacts:

ai_architect_github_create_pr(
  title="[{findingID}] {finding_summary}",
  body={assembled_pr_body},
  head="pipeline/{findingID}",
  base="main"
)

PR body template:
  ## Finding
  - **ID:** {findingID}
  - **Impact Score:** {compoundImpactScore}
  - **Source:** {finding_source}

  ## PRD Overview
  {prd-overview summary}

  ## Verification Summary
  - **HOR Rules:** {pass_count}/64 passed
  - **Build:** PASS
  - **Tests:** {passed}/{total} passed
  - **Compound Score:** {score}

  ## Audit Trail
  - Stage artifacts: `{data_dir}/artifacts/`
  - HOR report: `stage-7-hor-report.json`
  - Test report: `stage-9-test-report.json`
  - sha256 manifest: `sha256-manifest.json`

→ Returns PR URL

Read the full file on GitHub · 238 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 10d ago First seen · 238 lines · 4 tokens per session scan A fc5e7098ad3c

Subscribe to this mod's changes

stage-10-pr is a skill published in the GitHub repository cdeust/ai-architect-mcp (1 stars, last pushed 4mo ago), licensed MIT. It adds 4 tokens to every session and 2,269 once invoked, about $0.0000 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.