OpenHarness End-to-End Evals
HKUDS/OpenHarness
Validates OpenHarness features by running real multi-turn agent loops with live LLM calls against an unfamiliar codebase, checking actual tool execution.
Benchmarks how much CodeGraph helps a coding agent on a real repository, comparing runs with and without it for a chosen local or published version.
Install CodeGraph Agent Eval
$ npx skills add colbymchenry/codegraph --skill agent-eval -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install colbymchenry/codegraph agent-eval --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/colbymchenry/codegraph.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude/skills/agent-eval .claude/skills/agent-eval && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "agent-eval" agent skill from https://github.com/colbymchenry/codegraph/tree/main/.claude/skills/agent-eval into .claude/skills/agent-eval/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "agent-eval", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/colbymchenry/codegraph/tree/main/.claude/skills/agent-evalType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add colbymchenry/codegraph --skill agent-eval -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install colbymchenry/codegraph agent-eval --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/colbymchenry/codegraph.git skills-src && mkdir -p .agents/skills && cp -r skills-src/.claude/skills/agent-eval .agents/skills/agent-eval && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "agent-eval" agent skill from https://github.com/colbymchenry/codegraph/tree/main/.claude/skills/agent-eval into .agents/skills/agent-eval/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "agent-eval", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add colbymchenry/codegraph --skill agent-eval -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install colbymchenry/codegraph agent-eval --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/colbymchenry/codegraph.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/.claude/skills/agent-eval .cursor/skills/agent-eval && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "agent-eval" agent skill from https://github.com/colbymchenry/codegraph/tree/main/.claude/skills/agent-eval into .cursor/skills/agent-eval/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "agent-eval", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/colbymchenry/codegraph.git --path .claude/skills/agent-eval--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add colbymchenry/codegraph --skill agent-eval -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install colbymchenry/codegraph agent-eval --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/colbymchenry/codegraph.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/.claude/skills/agent-eval .gemini/skills/agent-eval && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "agent-eval" agent skill from https://github.com/colbymchenry/codegraph/tree/main/.claude/skills/agent-eval into .gemini/skills/agent-eval/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "agent-eval", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install colbymchenry/codegraph agent-evalInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add colbymchenry/codegraph --skill agent-eval -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/colbymchenry/codegraph.git skills-src && mkdir -p .github/skills && cp -r skills-src/.claude/skills/agent-eval .github/skills/agent-eval && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "agent-eval" agent skill from https://github.com/colbymchenry/codegraph/tree/main/.claude/skills/agent-eval into .github/skills/agent-eval/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "agent-eval", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add colbymchenry/codegraph --skill agent-eval -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install colbymchenry/codegraph agent-eval --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/colbymchenry/codegraph.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/.claude/skills/agent-eval .opencode/skills/agent-eval && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "agent-eval" agent skill from https://github.com/colbymchenry/codegraph/tree/main/.claude/skills/agent-eval into .opencode/skills/agent-eval/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "agent-eval", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
agent-eval- GitHub stars
- 73k
- Token cost
- ~950 tokens
- SKILL.md length
- 435 words
- Files
- 2
- Skills in repo
- 2
- Repo updated
- First seen
- Licence
- MIT
At a glance
Benchmarks how much CodeGraph helps a coding agent on a real repository, comparing runs with and without it for a chosen local or published version.
- Benchmarking a local CodeGraph build before a release
- SKILL.md covers Prerequisites, Workflow and Notes
- Calls claude
- Validating a published CodeGraph version against a real repo in one language
What it does
This skill drives a test harness in `scripts/agent-eval/` and is meant to be run from the codegraph repository root. It asks you, one question at a time, which CodeGraph version to test (the local dev build, the latest published release or a version you type), which language to cover, which repository from the `corpus.json` list (labeled with its size and file count) and which harness to use.
The harness is either headless, using `claude -p` with stream-json output for exact token and cost figures, or interactive, driving the real Claude interface inside tmux and reading metrics from session logs. A third option runs both. The audit script clones the repository if needed, wipes and re-indexes it, and runs in the background for several minutes.
When the job finishes, the agent reads the log and reports per run: tool calls, file reads, grep and bash calls, CodeGraph tool calls, duration and total cost for headless runs, and the verdict and token lines for interactive runs. A headless A/B run ends with a side-by-side comparison table.
When your agent uses it
- Benchmarking a local CodeGraph build before a release
- Validating a published CodeGraph version against a real repo in one language
- Comparing agent tool calls and cost with and without CodeGraph
Example prompts
- “Run the agent-eval audit on the latest published CodeGraph version and show me the arm comparison.”
- “Benchmark the local dev build of CodeGraph against a medium-sized repo from the corpus.”
- “Test whether CodeGraph 0.7.10 cuts down file reads on a larger codebase.”
Requirements
- `tmux` 3 or newer
- A logged-in `claude` CLI
- Node.js and Git on macOS or Linux
- The codegraph repository as the working directory
What it can do on your machine
Read from SKILL.md and the folder at commit b635dd4. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
Shell commands in SKILL.md call:
claude
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
CodeGraph Agent Eval loads about 950 tokens when it runs. Until then it costs about 75 tokens; SKILL.md has 435 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
SKILL.md
The full file from colbymchenry/codegraph at commit b635dd4, republished under its MIT licence (© colbymchenry). 435 words, ~950 tokens.
Download SKILL.mdSave it as
.claude/skills/agent-eval/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.- name
- agent-eval
- description
- Benchmark CodeGraph retrieval quality on a real codebase by comparing agent behavior with vs without CodeGraph. Use when the user runs /agent-eval or asks to test, benchmark, audit, or validate a codegraph version (the local dev build or a published npm version) against a language's repo.
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
tmux3+, a logged-inclaudeCLI,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 resultsStep 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" →
headless—claude -pwith 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, fileReads, Grep/Bash, codegraph-tool calls, duration, total cost. - Interactive (
parse-session.mjs): theVERDICT: codegraph_explore used Nx | Read N | Grep/Bash NandTOKENS:lines. - Both paths also print the three feedback metrics — residual context occupancy,
explore sufficiency, allocation efficiency — and a headless A/B ends with a
side-by-side
ARM COMPARISONtable. Report that table, and check its contamination row first:CLI calls that RETURNED output> 0 means the arm reached codegraph through Bash and its numbers are void. How to read the rest:docs/benchmarks/agent-eval-feedback-metrics.md.
Show full SKILL.md (100 more words)Show less
Lead with cost + tool/Read counts — they are the reliable signals; raw token in/out are confounded by subagent delegation and prompt caching. State whether codegraph reduced effort and whether both arms reached a correct answer.
Notes
- The index is rebuilt every run (
audit.shwipes.codegraph) — different versions extract differently, so an index must be served by the same binary that built it. audit.shtemporarily mutates the globalcodegraphinstall for the test, then restores your dev link vialocal-install.sh.- Corpus repos are cloned to
/tmp/codegraph-corpus(reused if already present). - Add or edit repos in
corpus.json(fields:name,repo,size,files,question).
© colbymchenry, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
SKILL.md and 1 other file in .claude/skills/agent-eval of colbymchenry/codegraph.
- SKILL.md
- corpus.json
Open the folder on GitHubat commit b635dd4
Compare with similar skills
CodeGraph Agent Eval next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| CodeGraph Agent Eval this skillcolbymchenry/codegraph | 73k | — | ~950 | Automated safety check: Pass | MIT | |
| OpenHarness End-to-End EvalsHKUDS/OpenHarness | 16k | 1 repos | ~2.1k | Automated safety check: Notes | MIT | |
| Release Clawpatchopenclaw/clawpatch | 813 | — | ~1.1k | Automated safety check: Pass | MIT | |
| Install Rhomikeyobrien/rho | 372 | — | ~1.6k | Automated safety check: Notes | MIT | |
| Darwin Skill Optimizeralchaincyf/darwin-skill | 6.2k | 1 repos | ~4.7k | Automated safety check: Pass | MIT | |
| Huashu Agent Swarmalchaincyf/huashu-skills | 1.7k | 1 repos | ~576 | Automated safety check: Pass | MIT |
Similar skills
Release Clawpatch
openclaw/clawpatch
clawpatch release: version/changelog, CI, npm publish, GitHub release, verify.
Install Rho
mikeyobrien/rho
Install and configure Rho from scratch (Doom-style init.toml + sync).
Darwin Skill Optimizer
alchaincyf/darwin-skill
Scores SKILL.md files on a nine-dimension rubric, then improves them in a keep-or-revert loop with independent judge agents, test prompts, git history and human checkpoints.
Huashu Agent Swarm
alchaincyf/huashu-skills
多Agent蜂群并行协作，纯git自组织，适合大型项目开发。当用户提到"蜂群模式"、"多agent"、"并行开发"、"agent swarm"时使用。
Agent Deck
asheshgoplani/agent-deck
agent-deck, the terminal session manager for AI coding agents.
More from colbymchenry/codegraph
Add a Language to CodeGraph
colbymchenry/codegraph
Adds tree-sitter support for a new language to the codegraph project, then tests it and benchmarks extraction quality and retrieval value on real repositories.
Related
Categories
Questions about CodeGraph Agent Eval
What does CodeGraph Agent Eval do?
Benchmarks how much CodeGraph helps a coding agent on a real repository, comparing runs with and without it for a chosen local or published version. This skill drives a test harness in `scripts/agent-eval/` and is meant to be run from the codegraph repository root.json` list (labeled with its size and file count) and which harness to use.
When should I use CodeGraph Agent Eval?
CodeGraph Agent Eval fits situations like: benchmarking a local CodeGraph build before a release; validating a published CodeGraph version against a real repo in one language; comparing agent tool calls and cost with and without CodeGraph.
How do I install CodeGraph Agent Eval in Claude Code?
Run `npx skills add colbymchenry/codegraph --skill agent-eval -a claude-code`. Or copy the skill folder (.claude/skills/agent-eval in colbymchenry/codegraph) into .claude/skills/agent-eval in your project. Claude Code loads it when a task matches its description.
How do I install CodeGraph Agent Eval in Codex?
Run `npx skills add colbymchenry/codegraph --skill agent-eval -a codex`. Or copy the skill folder (.claude/skills/agent-eval in colbymchenry/codegraph) into .agents/skills/agent-eval in your project. Codex loads it when a task matches its description.
Can I use CodeGraph Agent Eval in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add colbymchenry/codegraph --skill agent-eval -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/agent-eval, .gemini/skills/agent-eval, .github/skills/agent-eval and .opencode/skills/agent-eval in your project.
What does CodeGraph Agent Eval need to run?
Going by SKILL.md and its folder, CodeGraph Agent Eval needs the command-line tools its instructions call (claude). Our summary lists: `tmux` 3 or newer; A logged-in `claude` CLI; Node.js and Git on macOS or Linux; The codegraph repository as the working directory.
Does CodeGraph Agent Eval access the network?
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Is CodeGraph Agent Eval safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
What licence does CodeGraph Agent Eval use?
CodeGraph Agent Eval is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does CodeGraph Agent Eval use?
About 950 tokens (SKILL.md is roughly 3.8k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
What are the alternatives to CodeGraph Agent Eval?
Skills that share tags, products or a category with CodeGraph Agent Eval: OpenHarness End-to-End Evals (HKUDS/OpenHarness, 16k stars), Release Clawpatch (openclaw/clawpatch, 813 stars), Install Rho (mikeyobrien/rho, 372 stars) and Darwin Skill Optimizer (alchaincyf/darwin-skill, 6.2k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains CodeGraph Agent Eval?
colbymchenry (a GitHub user) maintains it in colbymchenry/codegraph, which has 73,427 GitHub stars. The repository holds 2 skills in this directory. The repository was last updated on October 7, 2026.
Source: colbymchenry/codegraph on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.