Merge two open `<issue-tracker>` issues that describe the same root cause, preserving both reporters' context. Proposes a closing comment on the duplicate and a cross-reference comment on the kept issue. Waits for maintainer confirmation before posting anything or closing anything.

Draft one net-new *good first issue* on the configured `<upstream>` repo from one supplied gap or small maintainer-named task. Run suitability and readiness checks before showing the draft. File via `gh` only after explicit maintainer confirmation. Never curate or relabel the existing backlog.

Walk an adopter through the first-time install of the secure agent setup (sandbox, approval and clean-environment layers) for Claude Code, Codex, Antigravity, Pi or Goose. Interactive throughout; never runs sudo, edits a shell rc, or overwrites settings on its own.

Import new `<security-list>` reports into `<tracker>`: find threads not yet tracked, propose the imports (default: import unless rejected), create each tracker in `Needs triage`, and draft a receipt reply to the reporter. First step of the handling process.

Open a tracker for a security-relevant fix that already exists as a public `<upstream>` PR, with no `<security-list>` report. The tracker lands in `Assessed` with scope, PR-state and remediation fields filled from the PR; pairs with `security-cve-allocate`.

Sub-skill for reports relayed onto `<security-list>` by a broker (the ASF security team, a disclosure platform, a SOC) rather than sent by the reporter. Detects the relay, extracts the credit and the reporter-addressing rules through the adapters in `forwarders.enabled`, and hands the routing back. Never mutates the tracker.

**Source page:** [Your first skill](../your-first-skill.md) **Estimated time:** 60 minutes (40 min reading + 20 min exercises and self-check) **Lesson in sequence:** 4 of 11

**Source page:** [Debugging a skill](../debugging-skills.md) **Estimated time:** 50 minutes (30 min reading + 20 min exercises and self-check) **Lesson in sequence:** 6 of 11

Make an existing framework skill leaner without changing its behavior. Diagnose context-cost smells, propose the applicable optimization passes, and validate before and after every approved change.

Paste everything below the line into the system prompt field of any capable chat model (Claude, GPT, a local model, etc.). The learner then talks to it in the normal chat window. Nothing above the line is sent to the model.

Draft the `[VOTE]` email body and planning-issue comment for an RC of `<upstream>`. Reads RC metadata from the planning issue and `<project-config>/release-management-config.md`; produces a ready-to-copy `[VOTE]` subject + body and a proposed planning-issue comment. Never sends mail and never posts without explicit RM confirmation.

Show a Magpie welcome guide on demand: the general one with no argument, or a skill family's by name. Also turns a guide back on after it was dismissed or hidden. Read-only apart from that.

Read-only review of one repository's GitHub Actions workflows and recent run history against the ASF Infra workflow-optimization practices. Produces a prioritised plan to cut runner use; never edits workflows, opens PRs, or posts comments.