farty-bobo

Post or update a comment on a Linear ticket using the Linear MCP server. Accepts a ticket ID (e.g. ENG-123), a Linear ticket URL, or a direct link to an existing comment. Works for both human-authored comments and agent-generated summaries.

Creates a Linear document (the Documents feature, distinct from issues/projects) from context in the current conversation, a plan, or a file. Use when the user wants to write up a doc, spec, RFC, or notes page in Linear and attach it to a team, project, initiative, cycle, or issue.

Used to run githooks, perform code & plan reviews with expert critics, and commit/push verified changes

Creates a Linear project under an initiative from a draft list of tickets, clarifies ambiguity with the human, then creates the project and each ticket via /create-linear-ticket. Use when the user wants to stand up a new project in Linear from a plan, doc, or ticket list.

Used to build and execute an approved implementation plan by spinning up agents

Used to gather requirements, clarify scope, and create implementation plans for new feature requests or bug fixes

Request an automated review on a GitHub PR. Prompts the human to pick a reviewer (Copilot, Claude, Codex on GitHub, or Codex within Claude Code). Also runs the bot feedback loop: waits for CI/bot reviews, addresses comments, resolves CI failures, then marks draft PRs ready for review.

Creates a GitHub Gist from content in the current conversation or a file. Secret by default; pass --public to make it public.

Help users of the Yaz connector explore org designs, images, and product visuals. Accepts an org name, design query, or product filter and returns structured results with context. Also supports auditing image coverage and spotting gaps in product imagery.

Merges the PR related to the current task and transitions the related ticket to Done. Supports both Linear and Jira (Atlassian). Confirms all actions with the human before executing.

Code review a pull request or a set of related PRs created by someone else. Reads diffs, comments inline, and posts a consolidated review summary.

Review one or more pull requests by spinning up one review agent per PR, then consolidating findings into a single cross-PR summary. Use this when reviewing 1+ PRs that need to be understood together — stacked, parallel feature work, or a release batch.

Used to read PR comments on github and act on them

Investigate and triage a Honeybadger fault or error. Accepts a fault ID, fault URL, or project name. Pulls fault details, recent occurrences, affected users, and occurrence trends, then produces a structured triage summary with severity assessment and recommended next steps.

Breaks a large epic into sequenced child tickets and runs /plan-task on each one in order. Use when the user provides an epic (Jira ID, markdown file, or written description) that needs to be decomposed into individually planned pieces of work.

Creates a Linear issue from context in the current conversation or codebase. Use when the user wants to log a bug, feature request, task, or chore in Linear.

Given a task or user query, check whether an existing skill, hook, or script in the farty-bobo repo already handles it. Always run this before building anything new. Trigger on: "do we have a tool for", "is there a skill for", "can we do X already", "check if there's a tool", "find a tool for", "what skill should I use for".

Find your open PRs that are either approved or have new unresolved/unresponded comments, compile them into a triage board with human-annotatable MERGE/ADDRESS/REPLY/SKIP actions, then dispatch dedicated agents to execute each decision. Use when the user asks "what PRs do I need to deal with?", "triage my open PRs", "which PRs need my attention?", or similar.

Channels the comedic style of Lewis Black — exasperated, incredulous, ranty — while still being genuinely helpful. Use this skill whenever the user is visibly frustrated, angry, or using profanity in their message. Also trigger when the user explicitly asks for a "Lewis Black take," "rant about this," "give it to me angry," or references Lewis Black by name. Even mild frustration cues like "this is ridiculous," "are you kidding me," "what the hell," or "I can't believe this" should activate this skill. If the user's message contains swear words (damn, hell, shit, fuck, ass, etc.) directed at a situation or problem (not at Claude), that's a trigger. Do NOT trigger on clinical or neutral uses of profanity in quoted text or code.

Scan Slack, Confluence, Jira, and GitHub for comments and messages that directly mention or tag the human. Present findings for selection, draft responses in a triage file, let the human APPROVE/DISMISS/RETRY each one, then dispatch approved responses using the appropriate skills.

Create a Confluence page from any context — a GitHub PR, current session learnings, code exploration, architecture decisions, incident postmortem, or freeform input. Synthesizes content from multiple sources into a structured Confluence page.

Summarize findings or answer questions posed by someone on Slack using facts found in the code or tools available to Farty Bobo. Farty Bobo identifies itself, uses respectful language, and maintains character.

Add an MCP server to the current project by creating or updating a project-level .mcp.json, .env, and .gitignore

Selectively install skills, hooks, and commands from the Farty Bobo config repo (fartybobo/farty-bobo) onto this machine — without forking or cloning the whole repo. Trigger on: "install farty bobo", "install skills from farty bobo", "get the farty bobo skills", "install hooks", "install commands from config repo".

Delete dead git worktrees. A worktree is dead if it has no uncommitted changes, OR if its branch was pushed to origin and that branch has since been deleted from origin (e.g. after a merged/closed PR). Trigger when the user says things like "clean up worktrees", "delete dead worktrees", "prune old worktrees", "remove stale worktrees", or "worktree cleanup".

Post a footer or inline comment on a Confluence page using the Atlassian MCP server. Accepts a Confluence page URL or a direct comment URL. The comment is attributed to your identity as defined in CLAUDE.md.

Post or update a comment on a Jira ticket using the Atlassian MCP server. Accepts a ticket ID (e.g. PROJ-123), a Jira ticket URL, or a direct link to an existing comment. Works for both human-authored comments and agent-generated summaries.

Creates a Jira ticket from context in the current conversation or codebase. Use when the user wants to log a bug, feature request, task, or spike in Jira.

Report the user's own open pull requests that have already been approved by at least one reviewer. Use when the user asks "what PRs of mine are approved?", "what's ready to merge?", "which of my PRs got an approval?", or similar. Surfaces PRs that are waiting on the author to merge, address nits, or rebase.

Generates a bash script to rotate AWS IAM access keys for a given user. Use this skill whenever the user mentions rotating AWS keys, refreshing AWS credentials, replacing access keys, or asks about AWS key rotation best practices. Also trigger when the user says things like "my AWS keys are old", "time to rotate credentials", "new access key", or "update my AWS secrets". This skill reads ~/.aws/credentials to identify the current key and produces a safe, reviewable rotation script — it never executes AWS commands directly.

Use when reviewing code for security vulnerabilities, implementing authentication/authorization, handling user input, or discussing web application security. Covers OWASP Top 10:2025, ASVS 5.0, and Agentic AI security (2026).

Used to investigate and fix errors recorded by ci/cd for the current branch/PR