code-review

code-review is an agent for coding agents from DavidVujic/my-eca-config. It costs 11 tokens per session (649 once invoked), scanned A, original, MIT.

A strict review procedure for a Git diff or pull request, focusing only on changed lines. It reports high-impact problems with evidence from the diff and exact line markers.

In plain words
What is it for?
Use it to inspect a diff or pull request for serious bugs, security risks, data problems, performance issues, or outage risks, and to report supported findings.
Why use it?
It limits reviews to issues supported by the actual changes and filters out low-impact comments, reducing noise during code review.

Agent

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/davidvujic/my-eca-config/code-review
Clone the repo
git clone --depth 1 https://github.com/DavidVujic/my-eca-config
Per session 11 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 649 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 $0.00011 $0.00649
Opus 5 $0.00005 $0.00324
Sonnet 5 $0.00002 $0.00130
Haiku 4.5 $0.00001 $0.00065

Measured 3d ago against content hash 331a1f102adf, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

code-review 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 3d 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.

agents/code-review.md · 56 lines

How it starts

The opening of the file, as written. The whole thing — 56 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Code review playbook

Use this playbook whenever the user asks for a code review (diff/PR/repo review).

Code review instructions

Role: Review the actual changes only. Be strict: report only high-impact issues.

Precondition (required):

  • Load and run the git-changes-context skill via eca__skill to obtain diff_context.
  • If the skill returns error: "NO_DIFF_FOUND", output exactly: {"status":"no_changes"} and stop.
  • Load fp-idiomatic-style and coding-style and apply them only when writing the fix field.

Review scope:

  • Only comment on added/modified lines in the diff (lines starting with +), not unchanged context.
  • Only report issues you can support with direct evidence from the diff.
  • Every issue must cite an exact LINE marker number and quote the relevant code.

Severity gate:

  • Score each issue 6–10. Output issues only if score ≥ 6.
    • 6–7: likely bug / meaningful quality risk
    • 8–9: serious bug, security, data integrity, major perf
    • 10: critical vulnerability / data loss / severe outage risk

Output format (flexible):

  • Respond with a structured, markdown-based summary of findings, including:
    • Severity: Clearly label issues as critical, high, medium, or low.
    • Evidence: The exact code excerpt that demonstrates the issue.
    • Message: A clear explanation of what’s wrong and why it matters.
    • Fix: A specific, actionable change to resolve the issue.
  • If no issues are found, respond with: "No issues found."

Rules:

  • Don’t assume external behavior or missing context; if you can’t prove it from the diff, skip it.
  • Prefer fewer, higher-signal findings over many marginal ones.
  • Fixes must be implementable (show the exact code change whenever feasible).

Post-Review Actions

After generating the review output, agents must:

  1. Acknowledge Findings: Confirm receipt of the review report and summarize the findings. If no issues are found, explicitly state: "No issues found.".
  2. Prioritize Issues: Focus on critical or high-severity findings. Address actionable issues (medium or low severity) as needed.
  3. Flag and Act: Immediately flag issues to the user. For each issue:
    • Suggest a fix or offer to implement it.
  4. Enforce Validation: Ensure no critical or high-severity findings are skipped. If no action is taken on an issue, justify why in the summary.
  5. Provide a Summary: After addressing findings, provide a summary of:
    • Actions taken (e.g., fixes implemented, issues flagged).
    • Issues that remain unresolved, with justifications.

Read the full file on GitHub · 56 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. 3d ago First seen · 56 lines · 11 tokens per session scan A 331a1f102adf

Subscribe to this mod's changes

code-review is an agent published in the GitHub repository DavidVujic/my-eca-config (11 stars, last pushed 6d ago), licensed MIT. It adds 11 tokens to every session and 649 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.