fix

An issue-fixing workflow that runs an eight-step debugging process on actionable defects, one issue at a time, and prepares each fix for verification.

In plain words
What is it for?
Use it to rank open bugs, create a development-branch fix, add tests and release notes, and submit the change for review.
Why use it?
It provides a controlled path from diagnosis to a tested pull request while preserving release safeguards and avoiding duplicate work.

Command for Claude Code

Install

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.

agentmods
npx agentmods add commands/littlebearapps/untether/fix
Clone the repo
git clone --depth 1 https://github.com/littlebearapps/untether

Made for: Claude Code.

Per session 76 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,556 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What 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.

ModelPer sessionOnce invoked
Fable 5 $0.00076 $0.02556
Opus 5 $0.00038 $0.01278
Sonnet 5 $0.00015 $0.00511
Haiku 4.5 $0.00008 $0.00256

Measured yesterday against content hash b9b9ec0a8342, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

fix 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 yesterday.

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.

.claude/commands/fix.md · 188 lines

How it starts

The opening of the file, as written. The whole thing — 188 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are handling the user's /fix request. /fix is Untether's action half of the debug engine: it runs the same 8-step protocol as /debug, but with ship authority — it sweeps the open actionable-defect list to a verified hand-off (PR to dev + needs-verification), one issue at a time, rank-ordered.

User input: $ARGUMENTS

Untether adaptations (read first)

Load .claude/rules/workflow-commands.md — the routing table + the 7 cross-cutting rules. In particular for /fix:

  • Release-guard obedience. Branch off dev; PR to dev; never push/merge to master, never tag, never gh release create, never --no-verify. The GitHub branch ruleset + CODEOWNERS is the real gate.
  • Dev/staging separation. Verify fixes on untether-dev.service; never restart untether.service (staging). Never systemctl restart from inside this session (hot-reload drain drops the final message).
  • Confirm-gated + idempotent. Surface each drafted PR body / issue comment and wait for a tap; de-dupe against open and closed issues; a re-invoked /fix must not double-open a PR or double-comment.
  • Redaction. Scrub tokens/keys/env/chat-content/fleet identifiers from any evidence before it lands in a PR or issue (see .claude/commands/debug/step-evidence.md).
  • Untether-mode. AskUserQuestion/ExitPlanMode return empty under Telegram — state assumptions in text and STOP for a reply. Keep the run report brief (≈500–1500 chars).

Boundary vs /debug

/debug = diagnose authority (pauses after hypothesis; ships only a minimal ≤3-file fix in targeted). /fix = the whole-list action sweep. One ship-path (.claude/commands/debug/step-fix.md); two authority levels. If a target needs investigation first, run /debug — don't guess a fix.

Sub-commands

Form Action
/fix (no args) Full sweep: rank the open actionable-defect list, process top-ranked to a hand-off
/fix #NN [#MM ...] Scoped: only the named issues, still rank-ordered
/fix --dry-run Propose-only: stage locally, print the would-be PR(s); open nothing
/fix --resume-from #NN Resume a sweep from a specific issue (skip already-handled)
/fix --help Print usage and stop

Read the full file on GitHub · 188 lines

Changes

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.

  1. yesterday First seen · 188 lines · 76 tokens per session scan A b9b9ec0a8342

Subscribe to this mod's changes

fix is a command published in the GitHub repository littlebearapps/untether (66 stars, last pushed 2d ago), licensed MIT. It adds 76 tokens to every session and 2,556 once invoked, about $0.0004 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 2026-08-30.