Engineering Weekly Decision Retro

    1

    Turn supplied git, ticket, incident, and planning exports into an evidence-bounded weekly manager memo—without connecting to systems or judging people.

    Free

    3 usesSecurity scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Engineering Weekly Decision Retro

    Example session with this skill installed

    Analyze these exports for week 42: git_commits.csv, jira_tickets.json, incidents.txt. Team: payments-platform. All exports are complete_for_window. window: 2023-10-16 to 2023-10-22.

    • Read your context and instructions
    • Compiled the engineering weekly decision

    Engineering Weekly Decision Retro — payments-platform — 2023-10-16 to 2023-10-22

    Evidence boundary

    • Evidence status: complete
    • Supplied exports: Git (12 commits), Tickets (5), Incidents (1)

    Plan drift (FACT)

    • completed_unplanned_item — Evidence: TICKET-789 (Done), no matching plan_id.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    The problem

    Engineering managers often waste hours manually correlating git commits, Jira tickets, and incident logs to prepare for weekly planning. Without a strict evidence-based approach, these retrospectives often devolve into subjective performance assessments or biased causality claims.

    What it does

    • Synthesizes git, ticket, and incident exports into a manager-ready weekly decision memo.
    • Flags planning drift by identifying completed unplanned work and unaddressed planned items.
    • Surfaces explicit friction signals like blocked tickets and open incident records.
    • Generates evidence-linked decision prompts for human review rather than automated mandates.
    • Proposes a single, reversible one-week experiment based on observed friction patterns.

    Frameworks & tools

    Works with structured exports (JSON, CSV, or text) from GitHub, GitLab, Jira, Linear, PagerDuty, and similar engineering ops tools.

    Why this beats prompting it yourself

    This skill enforces a strict read-only evidence contract that prevents the LLM from hallucinating causality or making inappropriate HR assessments. It handles the tedious work of record linkage and status normalization while maintaining a fail-closed gate for incomplete data.

    Use cases

    • Prepare a factual foundation for weekly engineering syncs or sprint reviews.
    • Identify systemic blockers and friction without resorting to individual blame.
    • Analyze plan-vs-actual drift using strictly bounded export windows.
    • Design low-risk process experiments based on concrete incident and ticket data.

    Known limitations

    Does not connect to live APIs or repositories. Author names are excluded from analysis. Cannot infer causality or individual performance from incomplete exports.

    Requires manual exports for input. No live system access. Suppresses trend analysis if data is marked partial. Does not assess individual developer performance.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    3 uses

    Times used to date

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 3 months ago

    • Free with an Agensi account

    Permissions

    No special permissions declared or detected

    Listed3 months ago

    What's inside

    Frequently Asked Questions

    No, this skill operates on a strict 'evidence-bounded' principle using manual exports (JSON/CSV) to ensure data privacy and prevent unauthorized system access.

    The skill is designed to be model-agnostic and works with most leading LLM agents that support file uploads or long-context text input.

    It focuses on system friction and 'unplanned vs. planned' drift, intentionally excluding author names to prevent subjective performance judging or HR-related bias.

    You will receive the specialized agent configuration, data formatting templates for your exports, and a guide on generating the weekly decision memos.

    Unlike basic prompting, this skill uses a structured framework to normalize different data sources and includes a 'fail-closed' gate to prevent the AI from hallucinating causality when data is missing.

    Updates are released periodically to support new export formats from tools like Linear or PagerDuty, and buyers receive lifetime access to these configuration refinements.

    New in Business & Operations