A workflow for reviewing existing pull requests and deciding whether they can be merged. A pull request is a proposed set of code changes awaiting review.

Use it to adopt specified pull requests, run their review checks, monitor CI, coordinate confirmed fixes, and merge only approved changes.

It organizes repeated review, testing, issue-fixing, and merge checks so a pull request is not merged before its required gates pass.

campaign — Skill by lestrrat-ai

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 skills/lestrrat-ai/claude-code-plugins/campaign

Any agent

npx skills add lestrrat-ai/claude-code-plugins --skill campaign

Clone the repo

git clone --depth 1 https://github.com/lestrrat-ai/claude-code-plugins

Made for: Claude Code, Codex.

Per session 183 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 9,860 The whole file, excluding the scripts and references it only reads on demand.

Security scan C 1 finding. 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.

Model	Per session	Once invoked
Fable 5	$0.00183	$0.09860
Opus 5	$0.00092	$0.04930
Sonnet 5	$0.00037	$0.01972
Haiku 4.5	$0.00018	$0.00986

Measured 2d ago against content hash dd0ab3e976df, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade C, and why

campaign scanned grade C with 1 finding 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.

The scan reads SKILL.md. This mod also ships 35 executable files (scripts/_gauntlet/__init__.py, scripts/_gauntlet/argv.py, scripts/_gauntlet/atomic.py, …), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

Recursive force deletehighDestructive command

rm -rf with a variable or a broad path is one typo away from removing the wrong tree.

- NEVER `rm -rf .gauntlet/`; only `.gauntlet/tmp/**` is disposable — **everything else under

plugins/gauntlet/skills/campaign/SKILL.md · 426 lines

How it starts

The opening of the file, as written. The whole thing — 426 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Campaign

Self-looping, reactive PR-review-to-merge pipeline. The active host is orchestrator + gatekeeper: reviews, CI watches, and fixes run as background tasks — and the heartbeat reconcile in a fresh synchronous worker where the host provides one; gates and merges stay centralized. Campaign gates existing PRs — adopted, never generated — and never writes a fix from scratch. To find issues first, use gauntlet:review; after its report it can open one PR per confirmed fix and hand them here (/gauntlet:campaign #PRs in Claude Code, $gauntlet:campaign #PRs in Codex).

Invoke once: the skill drives its own loop through the active host's heartbeat or bounded-wait mechanism. In Claude Code, do not wrap it in /loop; in Codex, keep the invocation alive when no heartbeat scheduler is available.

At every entry/resume, before any other work:

Read references/runtime-adapter.md. Resolve the supplied checkout through its typed RepositoryContext owner exactly once per invocation/resume, and carry that record for every repository path and Git cwd. The adapter owns every host mapping — invocation forms, model classes (session/economy), the heartbeat mechanism — and the typed process/data boundary: dynamic values cross as argv, byte-file, or native-message data, and each review attempt's record assigns exactly one final-report producer.
Read references/run-identity-and-lease.md and references/files-and-ledger.md before touching run state.
Read references/startup.md before starting or resuming a run whose pending_adoption checkpoint is still set. Its command protocol owns fresh-run setup.

The adversarial reviewer is a selectable role: by default the cross-engine route (Claude Code reviews with codex exec, Codex reviews with claude -p), launched at native-limitation level whenever the paired CLI is present, falling back to a fresh native worker only when it is absent or the reviewer is genuinely unusable. An explicit invocation or a TRUSTED saved preference (the orchestrator's own out-of-checkout user memory / global user instructions — NEVER a file inside the candidate checkout, including .gauntlet/history/ carryover) overrides the default. references/reviewer.md owns selection and fallback; references/runtime-adapter.md owns the isolation contract. Every route guarantees fresh conversational context and launches on that alone; installed campaign rules remain the stage-0 gate authority.

Read the full file on GitHub · 426 lines

Files

What ships with it

60 files 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.

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.

2d ago First seen · 426 lines · 183 tokens per session scan C dd0ab3e976df

Subscribe to this mod's changes

campaign is a skill published in the GitHub repository lestrrat-ai/claude-code-plugins (18 stars, last pushed 3d ago), licensed MIT. It adds 183 tokens to every session and 9,860 once invoked, about $0.0009 per session on Opus 5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Code review DevOps and cloud Git and workflow

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

brainstorming

You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.

obra/superpowers · 37 tokens

chat-pet-sprite-creation

Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.

microsoft/vscode · 53 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens

agent-host-chat-contributions

Build and review cross-cutting agent-host chat behavior through lifecycle contributions. Use when adding turn lifecycle side effects, prompt or context injection, restored-history transformation, protocol-action observation, or when reviewing changes that add code to AgentSideEffects or AgentService.

microsoft/vscode · 56 tokens

auto-perf-optimize

Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.

microsoft/vscode · 62 tokens

campaign