repair

repair is a command for Claude Code from theshadow27/mcp-cli. It costs 0 tokens per session (1,941 once invoked), scanned A, original, MIT.

A workflow for fixing issues reported against an already-open GitHub pull request. The findings may come from code review, QA, automated checks, or tests.

In plain words
What is it for?
Use it to verify the repository and branch, apply listed fixes, push the changes, reply to open review threads with the fix commit, and prepare the pull request for re-QA.
Why use it?
It turns review comments and failures into tracked fixes and leaves the pull request ready for another QA check without merging it.

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/theshadow27/mcp-cli/repair
Clone the repo
git clone --depth 1 https://github.com/theshadow27/mcp-cli

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for repair

README.md
[![agentmods](https://agentmods.dev/badge/commands/theshadow27/mcp-cli/repair.svg)](https://agentmods.dev/commands/theshadow27/mcp-cli/repair)
Your own site
<a href="https://agentmods.dev/commands/theshadow27/mcp-cli/repair"><img src="https://agentmods.dev/badge/commands/theshadow27/mcp-cli/repair.svg" alt="Measured on agentmods" height="20"></a>
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,941 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 $0.00000 $0.01941
Opus 5 $0.00000 $0.00971
Sonnet 5 $0.00000 $0.00388
Haiku 4.5 $0.00000 $0.00194

Measured 4d ago against content hash b30e86e07fa6, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

repair 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 4d 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.

.claude/commands/repair.md · 233 lines

How it starts

The opening of the file, as written. The whole thing — 233 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Repair an Open PR

You are a developer fixing issues raised against an already-open PR. The fixes are usually small, contained, and come from one of four sources:

  1. Copilot inline review comments on the PR
  2. QA blockers posted by a previous /qa session that labeled qa:fail
  3. Adversarial review findings posted as a sticky review comment
  4. CI failures (lint, typecheck, test, coverage)

Your job is to apply the fixes, push, reply to each open thread citing the fix commit, and leave the PR ready for re-QA. You do not merge the PR — the orchestrator owns that.

Input

The orchestrator will provide: PR number, branch name, and a bulleted fix list. Parse from: $ARGUMENTS

Typical prompt shape:

/repair PR #1472 (branch feat/issue-1441-worktree-containment).

Address these N findings:
1. containment.ts:143 (Copilot comment id 3106183084) — <description of the fix>
2. containment.ts:155 (Copilot comment id 3106183090) — <description>
3. ws-server.ts:1556 (id 3106183097) — <description>

Workflow

Step 0: Verify context

Confirm you're in the right worktree on the right branch before touching any code. Repair sessions are typically spawned with --worktree and should already be checked out to the PR branch — but verify, don't assume.

gh pr view <pr-number> --json number,headRefName,state,mergeStateStatus
git rev-parse --abbrev-ref HEAD    # must match headRefName above
git fetch origin <branch>
git pull origin <branch>            # rebase onto latest pushed state

If the branch is wrong, stop and report to the orchestrator. Don't git checkout on a shared main checkout — that's the escape pattern caught by the containment guard (#1441, #1425, #1488). If the PR is closed or merged, stop and report.

Step 1: Collect the open threads

Read every unresolved comment on the PR. A "fix list" in the prompt is the starting point, not the complete set — sometimes new comments have landed since the orchestrator wrote the prompt.

Read the full file on GitHub · 233 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. 4d ago First seen · 233 lines · 0 tokens per session scan A b30e86e07fa6

Subscribe to this mod's changes

repair is a command published in the GitHub repository theshadow27/mcp-cli (2 stars, last pushed 5d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,941 tokens. 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.