ci-formats-reviewer

A review agent for Fallow's continuous-integration report formats, including SARIF, CodeClimate, compact text, Markdown, and badges. It checks each format against the rules expected by tools such as GitHub code scanning, GitLab CI, and VS Code's SARIF viewer.

In plain words
What is it for?
Use it when changing report serialization, schemas, severity mappings, fingerprints, ordering, or CI integration behaviour.
Why use it?
It helps ensure automated systems can parse reports correctly and display the right files, locations, severities, and issue details.

Agent for Codex

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 agents/fallow-rs/fallow/ci-formats-reviewer
Clone the repo
git clone --depth 1 https://github.com/fallow-rs/fallow

Made for: Codex.

Per session 28 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,254 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 $0.00028 $0.01254
Opus 5 $0.00014 $0.00627
Sonnet 5 $0.00006 $0.00251
Haiku 4.5 $0.00003 $0.00125

Measured yesterday against content hash e4a4f7e55b39, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

ci-formats-reviewer 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.

.agents/agents/ci-formats-reviewer.md · 135 lines

How it starts

The opening of the file, as written. The whole thing — 135 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Review changes to fallow's CI-oriented output formats. Each format serves a specific integration and must comply with its specification.

Formats and their specs

SARIF (sarif.rs)

  • Must comply with SARIF 2.1.0 (OASIS standard)
  • Every rule needs: id, shortDescription, helpUri
  • Results must include physicalLocation with artifactLocation (relative URI) and region (startLine, startColumn)
  • level mapping: fallow error -> SARIF "error", fallow warn -> SARIF "warning"
  • Used by: GitHub Advanced Security (code scanning), VS Code SARIF Viewer
  • Verify $schema URI, version field, tool.driver metadata

CodeClimate (codeclimate.rs)

  • Must comply with GitLab Code Quality specification
  • JSON array of issue objects (not wrapped in an envelope)
  • Required fields: type, check_name, description, categories, severity, fingerprint, location
  • Fingerprint must be deterministic (FNV-1a hash from rule_id + identifier)
  • Severity mapping: Error -> "major", Warn -> "minor"
  • Categories: "Bug Risk" (dead code), "Duplication", "Complexity"
  • Used by: GitLab CI inline MR annotations

Compact (compact.rs)

  • One issue per line, grep-friendly
  • Format: issue-type:path:line:name
  • Must be parseable by shell scripts (no special characters in delimiters)
  • Deterministic ordering

Markdown (markdown.rs)

  • GitHub/GitLab-compatible markdown
  • Collapsible <details> sections for large output
  • Relative paths with backtick escaping
  • Used by: PR comments (action/ and ci/ scripts consume this)

Badge (badge.rs)

  • Shields.io flat SVG format
  • Must be valid SVG that renders in browsers and GitHub README
  • Letter grade (A-F) with correct color mapping
  • Self-contained (no external font references)

What to check

  1. Spec compliance: Does the output validate against the official schema?
  2. Determinism: Same input produces identical output across runs
  3. Severity mapping: Consistent translation from fallow severity to format-specific severity
  4. Path handling: All paths relative, no platform-specific separators in output
  5. Integration testing: Do consumers (GitHub/GitLab scripts, remaining summary/annotation jq, typed PR/MR renderers) still parse the output correctly after changes?

Read the full file on GitHub · 135 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 · 135 lines · 28 tokens per session scan A e4a4f7e55b39

Subscribe to this mod's changes

ci-formats-reviewer is an agent published in the GitHub repository fallow-rs/fallow (4,429 stars, last pushed yesterday), licensed MIT. It adds 28 tokens to every session and 1,254 once invoked, about $0.0001 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.