agent-eval

A benchmark for measuring whether CodeGraph, a code-search and retrieval tool, helps an AI coding agent understand a real codebase better than ordinary file search and reading.

In plain words
What is it for?
Use it to test CodeGraph on repositories in a chosen programming language, compare retrieval results, and report the audit findings.
Why use it?
It replaces guesswork with a repeatable comparison of agent performance with and without CodeGraph. This helps validate a local or published CodeGraph version.

Skill for Claude CodeCodex

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/lagrahhn/codegraph-branch/agent-eval
Any agent
npx skills add lagrahhn/codeGraph-branch --skill agent-eval
Clone the repo
git clone --depth 1 https://github.com/lagrahhn/codeGraph-branch

Made for: Claude Code, Codex.

Per session 65 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 911 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin 92% copy Near-identical to another mod 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 $0.00065 $0.00911
Opus 5 $0.00032 $0.00456
Sonnet 5 $0.00013 $0.00182
Haiku 4.5 $0.00006 $0.00091

Measured 2d ago against content hash b83c65193e06, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

agent-eval 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 2d 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.

Origin

This is a copy

92% identical to agent-eval — 6 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

.claude/skills/agent-eval/SKILL.md · 75 lines

How it starts

The opening of the file, as written. The whole thing — 75 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CodeGraph Quality Audit

Measures how much CodeGraph helps an agent versus plain grep/read, for a chosen codegraph version on a chosen real-world repo. Drives the harness in scripts/agent-eval/.

Prerequisites

  • tmux 3+, a logged-in claude CLI, node, git (macOS/Linux).
  • Run from the codegraph repo root.

Workflow

Copy this checklist:

- [ ] 1. Pick version (local or npm)
- [ ] 2. Pick language
- [ ] 3. Pick repo by size
- [ ] 4. Pick harness (headless / tmux / both)
- [ ] 5. Run audit.sh in the background
- [ ] 6. Report results

Step 1 — version. Ask with AskUserQuestion: which codegraph version to test. Offer "Local dev build" and "Latest published"; the free-text "Other" lets the user type a specific version (e.g. 0.7.10). Map the answer to a VERSION token:

  • "Local dev build" → local
  • "Latest published" → latest
  • a typed version → that string (e.g. 0.7.10)

Step 2 — language. Read .claude/skills/agent-eval/corpus.json. Ask with AskUserQuestion which language to test, listing the languages that have entries.

Step 3 — repo. From the chosen language's entries, ask which repo. Label each option with its size and file count, e.g. excalidraw — Medium (~600 files). Each entry carries the repo URL and a representative question.

Step 4 — harness. Ask with AskUserQuestion which harness to run, and map the answer to a MODE token:

  • "Headless" → headlessclaude -p with stream-json: exact tokens/cost and a clean tool sequence (2 runs, fast, no TTY).
  • "Interactive (tmux)" → tmux — drives the real Claude TUI in tmux: faithful Explore-subagent behavior, metrics from session logs (2 runs, slower).
  • "Both" → all — headless + interactive (4 runs).

Step 5 — run. Launch in the background (sets the version, clones if missing, wipes + re-indexes, runs the chosen arms — several minutes):

scripts/agent-eval/audit.sh <VERSION> <repo-name> <repo-url> "<question>" <MODE>

Step 6 — report. When the job finishes, read the log and report per arm:

  • Headless (parse-run.mjs): total tool calls, file Reads, Grep/Bash, codegraph-tool calls, duration, total cost.
  • Interactive (parse-session.mjs): the VERDICT: codegraph_explore used Nx | Read N | Grep/Bash N and TOKENS: lines.

Read the full file on GitHub · 75 lines

Files

What ships with it

1 file 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.

  1. 2d ago First seen · 75 lines · 65 tokens per session scan A b83c65193e06

Subscribe to this mod's changes

agent-eval is a skill published in the GitHub repository lagrahhn/codeGraph-branch (2 stars, last pushed 1mo ago), licensed MIT. It adds 65 tokens to every session and 911 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. It is 92% identical to agent-eval, differing in 6 lines, and is treated as a copy.

Related

Other skills, from other repositories

graph

Whole-codebase structural map of the codanna index as one self-contained HTML disc. Wedges per top-level module, concentric rings with hubs at the centre, hover or click a symbol to light its edge web, search, hide or highlight modules, brush a date range on the ribbon timeline (symbols dated by git blame), heatmap of…

bartolli/codanna · 131 tokens

x-ray

Deep codebase exploration using semantic search and relationship mapping. Use when you need to understand the current codebase.

bartolli/codanna · 25 tokens

brainctl

Unified agent memory CLI — read, write, search, and maintain the shared memory spine (brain.db). Use for persistent cross-session memory, knowledge graph, event logging, decisions, affect tracking, and consolidation.

TSchonleber/brainctl · 45 tokens

adding-a-language

How a new language enters Kivgraph's graph - the two routes and why one of them is almost always wrong, the four decisions that cannot be migrated later, the branches that decide what your language is without asking, and the gates that fail closed. Use when adding or extending language support, when touching…

Luqueee/kivgraph · 90 tokens

codestory-grounding

Use when an agent should ground a local repository with CodeStory before making source claims, planning edits, choosing tests, reviewing changes, or using broad retrieval evidence through the CodeStory plugin MCP.

TheGreenCedar/CodeStory · 44 tokens

running-tests

Cómo se ejecutan y se escriben los tests de Kivgraph - las tres suites, el tag ladybug que no se invoca a mano, qué se salta y por qué, y el smoke test del binario. Usar al correr tests, al añadirlos, ante un fallo de enlazado library 'lbug' not found, ante un SKIP inesperado, o antes de cerrar una tarea.

Luqueee/kivgraph · 94 tokens