driangle

List tasks with optional filters. Use when the user wants to see their tasks.

List tasks with optional filters. Use when the user wants to see their tasks.

Get the next recommended task to work on. Use when the user asks what to work on next or needs a task assignment.

Create a new task file following the taskmd specification. Use when the user wants to add a new task to the project.

Split a large task into smaller sub-tasks. Accepts a task ID, evaluates complexity, and creates sibling task files if warranted.

Update an existing task's fields (status, priority, title, tags, dependencies, etc.). Use when the user wants to modify a task's properties.

Create a new task file following the taskmd specification. Use when the user wants to add a new task to the project.

Split a large task into smaller sub-tasks. Accepts a task ID, evaluates complexity, and creates sibling task files if warranted.

Open a GitHub PR for the current branch. Generates a title and description from the diff, and respects the repo's PULL_REQUEST_TEMPLATE if one exists.

Split a large feature branch into multiple smaller PRs of max 20 files each. Use when the user wants to break up a big branch, split changes into stacked or parallel PRs, asks "how do I split this PR", mentions a branch has too many files, or wants a PR splitting strategy. Also triggers on phrases like "split this into PRs", "break up this branch", "chunk these changes", "PR plan", or "too many files in this PR".

Generate a concise PR description based on the changed files and their content. Focus on summarizing the key changes and their impact, rather than providing a detailed line-by-line explanation.

Generate a Slack-friendly triage report of open GitHub PRs for a team, author, or set of repos — pure-prompt variant of pr-review-report with no helper script. Use this skill whenever the user wants a PR review queue summary, a daily/standup PR digest, a reviewer triage list, or asks things like "what PRs does my team owe reviews on", "which of my PRs are waiting on review", "show me the team's open PRs", or "give me a PR status report for org X". The skill buckets each PR into exactly one of Needs attention / Ready to merge / In discussion / Awaiting review so reviewers can focus on what matters most.

Pick up the next task, execute it, verify it, mark it complete, and commit. Use when the user wants to work through tasks one at a time, or says 'do the next task', 'work on the next item', or invokes /work. Optionally accepts a task ID and/or custom instructions.

Generate a Slack-friendly triage report of open GitHub PRs for a team, author, or set of repos. Use this skill whenever the user wants a PR review queue summary, a daily/standup PR digest, a reviewer triage list, or asks things like "what PRs does my team owe reviews on", "which of my PRs are waiting on review", "show me the team's open PRs", or "give me a PR status report for org X". The skill buckets each PR into exactly one of Needs attention / Ready to merge / In discussion / Awaiting review so reviewers can focus on what matters most.

Scan a scope of code for unconventional, strange, questionable, undocumented, or opaque things — latent bugs, unexpected complexity, inconsistencies, surprising design choices, and library/API workarounds. Produces a prioritized reading list, not a refactor plan. Use whenever the user wants to 'review' a directory/module/branch, flag 'weird' or 'suspicious' code, find 'red flags', spot 'code smells', or just understand what's odd in a slice of the codebase they're about to touch.

Review a GitHub PR and present findings as author-addressed comments. Use when the user asks to review a PR, gives a PR URL or number, or asks for code review feedback on a pull request. Triggers on phrases like "review this PR", "look at this pull request", "code review", or any GitHub PR link.

Audit documentation coverage across CLI commands, web features, and configuration. Builds the CLI, discovers all commands/flags, checks web pages, and cross-references against docs/ and apps/docs/.

Perform a comprehensive codebase audit covering security, privacy, data integrity, architecture, and code quality. Use when the user wants to audit the codebase, check for security issues, or review code quality.

Generate skival suite.yaml files for evaluating and comparing AI agent configurations. Use when the user wants to create an eval suite, compare runners/models/skills/tool-access/environments, benchmark agent performance, or measure correctness/cost/speed across different AI setups.

Add a new easing function to the easefn library — implementation, tests, and documentation. Use this skill whenever the user wants to add, create, or implement a new easing function, curve, or interpolation method. Also trigger when the user says things like 'add easeInX', 'new easing', 'implement a spring easing', or 'I want to add a custom curve'. Even if they just name a function like 'easeInElastic' or describe a curve behavior, this skill applies.

Release the Go library by running checks, tagging, and pushing. Use when the user says /release-go <version> or asks to release the Go module.