Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/derio-net/super-fr/fr-debuggingnpx skills add derio-net/super-fr --skill fr-debugginggit clone --depth 1 https://github.com/derio-net/super-frWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00134 | $0.01390 |
| Opus 5 | $0.00067 | $0.00695 |
| Sonnet 5 | $0.00027 | $0.00278 |
| Haiku 4.5 | $0.00013 | $0.00139 |
Grade A, and why
fr-debugging scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 116 lines — stays where its author put it; the contents beside it link to each section on GitHub.
fr-debugging
superpowers:systematic-debugging, wrapped in isolation. The root-cause
investigation, the failing test, and the fix all happen in the isolation
workspace, so the base repo is never touched and the fix lands as a reviewed
PR — the debugging analogue of fr-brainstorming / fr-goal.
Announce at start: "I'm using fr-debugging to find the root cause in isolation."
This skill owns WHERE debugging happens (isolation) and the autonomy contract.
It delegates HOW — the Iron Law, the four phases, the supporting techniques —
to superpowers:systematic-debugging, invoked unchanged.
0. Isolation first — hard gate
Before ANY command — reproduction, evidence-gathering, instrumentation, reads included; "just check X first" never reorders this:
- Already inside an active fr isolation workspace? (e.g. a bug found
mid-fr-goal implementation) — reuse it. The failing test + fix land on
that feature's branch and ride its existing PR; a bug found while building a
feature must not spawn a competing PR. Confirm with
fr isolation status. - Cold start (standalone bug, no active workspace):
Name the branch for the bug now (fr isolation up --branch fix/<slug> [--profile <name>]fix/<slug>); the worktree, PR, and cleanup key off it. A cold-startfix/<slug>is cut from freshly-fetchedorigin/<default>(#322), so the fix is clean of whatever the base checkout is parked on — to debug on the current branch by intent, add--base HEAD. No devcontainer profile → HARD STOP — offer the fr-init interview; there is no unisolated fallback. (Under fr-goal: treat as a blocker — pause, fr-init, resume.)
From here on follow fr-isolation's exec-bridge discipline: read/edit files in
the worktree, run every command through fr isolation exec -- ….
1. Debug
Run superpowers:systematic-debugging as written — the Iron Law (no fixes
without root cause), the four phases (root cause → pattern → hypothesis →
implementation), the red flags, and the supporting techniques
(root-cause-tracing, defense-in-depth, condition-based-waiting) all
apply unchanged. Phase 4's failing test uses
superpowers:test-driven-development.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 116 lines · 134 tokens per session scan A b3291aeb7946
fr-debugging is a skill published in the GitHub repository derio-net/super-fr (1 stars, last pushed 2d ago), licensed MIT. It adds 134 tokens to every session and 1,390 once invoked, about $0.0007 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
release
Release a new version of BranchBox. Use when the user says "release", "new version", "cut a release", "publish release", or "tag version". Handles version bumping, changelog updates, quality checks, and GitHub release workflow.
branchbox-devcontainer-guardrails
Use when modifying BranchBox devcontainer/bootstrap workflows, compose templates, feature env-stash behavior, or manual E2E harness/release docs. Apply issue.
using-grove
Use when orchestrating a fleet of independent coding agents with Grove from OUTSIDE — isolated per-task workspaces (tmux plus git worktree, host or container) across model providers and harnesses. Triggers on: spinning up, steering, pausing, resuming or killing workspaces, grove create/fleet/show/message/tickets, the…
working-in-grove
Use when you are the coding agent running INSIDE a Grove workspace — a git worktree Grove created for one task, on the host or in a container. Triggers on: reporting or updating your task phase or progress, GROVEPHASEFILE or .grove/phase.json, grove phase, grovesetworkspacephase, naming this workspace or setting its…
configuring-grove
Use when a user wants to set up or change Grove's configuration, for a host/user or for a project. Covers the global user config, the committed project config, machine-local overrides, agents, init scripts, containers, tmux and peek tuning, the TUI theme, the daemon auth, notifications, and ticket providers (Gitea…
code-review
Automated code review assistant.