Run the agent review

Dispatch the index’s agent-review.yml: have every configured LLM review a commit on main and file one summary issue. Started with the standard GitHub CLI; the run is watched with gh run watch.

Skills it satisfies

A skill states the capability; this Tool is one way to exercise it. Each skill’s declared input and output contract follows, where it has one.

adversarial-code-review

Review a pull request an agent touched by building the strongest case that it is wrong, unsafe, or not what it claims, and record the verdict where the merge gate reads it: one merge-review/v1 file per head, on the merge-reviews state branch, never in the pull request itself. Covers who may review, what the whole-diff scope is, the RED FLAG categories and their evidence rule, the one rebuttal round, overrides, and how a per-head verdict composes with a merge train.

Source: cat-harness/sdlc/skills/sdlc-core/adversarial-code-review.md.

The skill declares no I/O contract (schemas/skills/adversarial-code-review/{input,output}.schema.json).

Inputs and outputs

Each port names a shared type by IRI, so two Tools for one skill can be checked for interchangeability.

porttyperequireddescriptionconstraint
inrepoRepoFullNameThe repository whose workflow is run. Default: the current one.string · pattern ^[A-Za-z0-9][A-Za-z0-9._-]*\\/[A-Za-z0-9][A-Za-z0-9._-]*$ · A forge repository’s full name, owner/name
inrefBranchThe branch to run it on. Default: the default branch.string · pattern ^[A-Za-z0-9][A-Za-z0-9._/-]*$ · A git branch name
outrunTextThe dispatched run; its outcome is read with gh run watch or gh run view.string · A short single-line human string. Not admissible as a command-line argument.

Processes and tasks

Steps this Tool can carry out, because each names a skill it satisfies:

processsteplanevia skill
Code change and reviewReview the changeCode revieweradversarial-code-review

How it is deployed and reached

  • Command line. From the repository root: gh workflow run agent-review.yml.
  • GitHub Actions. No workflow in this checkout runs its command.

As an API: its operation in the OpenAPI document and the MCP tool list of its layer.

installcli: gh
runtimegh
networkyes

When a host it reaches is refused:

host / errorremedy
github.comnone: The workflow runs on GitHub; with no host there is nothing to dispatch. The scripts it runs can still be run locally by their own keys.
when to choose it
whenRunning the agent-review.yml workflow on demand instead of waiting for its trigger, or finding which workflow a CI step belongs to.
limitsNeeds a token that may dispatch workflows (actions: write) on the repository.
costA full CI run on GitHub’s runners.

What it is made of

Its definition names no file it is made of.

Linked data

The Tool is published as its own JSON-LD document at its IRI, https://litlfred.github.io/folio-assistant/cat-harness/tools/ci-agent-review.jsonld, partOf its layer’s document. The node below is the definition this page was drawn from, and is embedded in the page as application/ld+json.

{
  "@context": "https://litlfred.github.io/folio-assistant/cat-harness/cat-harness.context.jsonld",
  "@id": "https://litlfred.github.io/folio-assistant/cat-harness/tools/ci-agent-review.jsonld",
  "@type": "https://litlfred.github.io/cat-harness/v0/ns/Tool",
  "partOf": "https://litlfred.github.io/folio-assistant/cat-harness/cat-harness.jsonld",
  "name": "ci-agent-review",
  "title": "Run the agent review",
  "description": "Dispatch the index's `agent-review.yml`: have every configured LLM review a commit on main and file one summary issue. Started with the standard GitHub CLI; the run is watched with `gh run watch`.",
  "install": {
    "cli": "gh"
  },
  "invoke": {
    "shell": "gh workflow run agent-review.yml"
  },
  "io": {
    "inputs": [
      {
        "name": "repo",
        "schema": "https://litlfred.github.io/folio-assistant/cat-harness/tool-types.schema.json#/$defs/RepoFullName",
        "description": "The repository whose workflow is run. Default: the current one.",
        "required": false,
        "arg": {
          "flag": "--repo"
        }
      },
      {
        "name": "ref",
        "schema": "https://litlfred.github.io/folio-assistant/cat-harness/tool-types.schema.json#/$defs/Branch",
        "description": "The branch to run it on. Default: the default branch.",
        "required": false,
        "arg": {
          "flag": "--ref"
        }
      }
    ],
    "outputs": [
      {
        "name": "run",
        "schema": "https://litlfred.github.io/folio-assistant/cat-harness/tool-types.schema.json#/$defs/Text",
        "description": "The dispatched run; its outcome is read with `gh run watch` or `gh run view`."
      }
    ]
  },
  "satisfies": [
    "adversarial-code-review"
  ],
  "requirements": {
    "runtime": [
      "gh"
    ],
    "network": true
  },
  "remedies": [
    {
      "host": "github.com",
      "none": "The workflow runs on GitHub; with no host there is nothing to dispatch. The scripts it runs can still be run locally by their own keys."
    }
  ],
  "selection": {
    "when": "Running the `agent-review.yml` workflow on demand instead of waiting for its trigger, or finding which workflow a CI step belongs to.",
    "limits": "Needs a token that may dispatch workflows (actions: write) on the repository.",
    "cost": "A full CI run on GitHub's runners."
  }
}

folio-assistant — code under the Apache License 2.0, documentation under CC BY 3.0. Generated schema and API references are rebuilt from source on every push to main.

deployed dev