Snag | Agent Skills Atlas

About this skill

Reactive flaw-remediation for JellyRock — fires the moment a defect, bug, gap, or regression surfaces mid-work, BEFORE you hack a first-draft fix. Triages the flaw by size (blast radius + judgment required, NOT line count) and routes it behind a confirm gate: trivial → fix now with zero ceremony; small/medium → a structured in-session fix carrying RCA gates (confirm root cause / weigh ≥2 fixes / right-size the artifact / regression+dogfood via npm run validate/test:scripts/test:tdd / human gate), optionally delegated to an Opus sub-agent to preserve main-thread context; large → escalate to tracked work (/log followup or /start-project). Recommend-then-confirm, flat disclosure, non-mandatory — self-demotes so it never taxes a 2-line fix. Reuses /focus's triage→route architecture but is distinct by TRIGGER: /focus is proactive session-start backlog triage; /snag is reactive triage of a flaw found mid-work. Use when something turns out broken and you want to handle it at the right size instead of reflexively patching. NOT for picking the next task (use /focus), and NOT for a known crash/CI/issue investigation that already has a dedicated flow (/runtime-triage, /ci-triage, /issue-triage).

Quick start

Copy the install prompt. Use the green button above this README.
Send it to your agent. Paste it into your coding agent’s chat and send.
Review the setup. Your agent checks the skill’s instructions, requirements, and installation scope.

Installation guide →

Review, license & listing details

Review the source before using this skill

This is an imported listing. We haven’t certified its safety, license, or compatibility. Skills may include executable scripts.

Automated security scan: Partial

Only SKILL.md was scanned. Supporting files and scripts were not checked. No safety grade has been assigned.

0 potential findings · Scanned 11/4/2026 · Scanner 0.2.1

Content SHA-256: cf5c33405a72bae109fbe2297e712e121d68b8a7643ce157c4860c7663d80c92

About scan coverage and limitations

Original listing metadata

Original name

snag

Original description

Reactive flaw-remediation for JellyRock — fires the moment a defect, bug, gap, or regression surfaces mid-work, BEFORE you hack a first-draft fix. Triages the flaw by size (blast radius + judgment required, NOT line count) and routes it behind a confirm gate: trivial → fix now with zero ceremony; small/medium → a structured in-session fix carrying RCA gates (confirm root cause / weigh ≥2 fixes / right-size the artifact / regression+dogfood via `npm run validate`/`test:scripts`/`test:tdd` / human gate), optionally delegated to an Opus sub-agent to preserve main-thread context; large → escalate to tracked work (`/log followup` or `/start-project`). Recommend-then-confirm, flat disclosure, non-mandatory — self-demotes so it never taxes a 2-line fix. Reuses `/focus`'s triage→route architecture but is distinct by TRIGGER: `/focus` is proactive session-start backlog triage; `/snag` is reactive triage of a flaw found mid-work. Use when something turns out broken and you want to handle it at the right size instead of reflexively patching. NOT for picking the next task (use `/focus`), and NOT for a known crash/CI/issue investigation that already has a dedicated flow (`/runtime-triage`, `/ci-triage`, `/issue-triage`).

About this listing

Category: Engineering
Reported license: GPL-2.0
Listing language (estimated): English
Review status: Not individually reviewed
Agent compatibility: Atlas installation supports Claude Code, Codex, and Cursor. Skill-specific dependencies and features are checked during setup.