eval-report
Skill24 starsChanged yesterdayMITView on GitHub
The short report format for any eval result in this repo. Load it whenever you report on an eval: after running primitives-eval or any harness under evals/, when asked "how did the eval go", "is it working", "what did the runs show", or when summarising runs you or a subagent just finished. Four parts, in a fixed order, under 150 words.
4 other projects use a version of this.
Adopted by
Most recently taken up by
- malloydata/publishertheir own version, first seen Nov 14, 2026
- agentscope-ai/OpenJudgetheir own version, first seen Oct 25, 2026
- modelscope/OpenJudgetheir own version, first seen Oct 9, 2026
- Terum-Inc/terum-skillstheir own version, first seen Sep 20, 2026
Other versions4 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 4
- agentscope-ai/OpenJudge★ 1.1ksame name
- modelscope/OpenJudge★ 1.1ksame name
- malloydata/publisher★ 133same name
- Terum-Inc/terum-skills★ 28same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository16 other files it gives its agents.
- AGENTS.mdAGENTS.md
- framework/agents.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- adversarial-testsSkill
- app-introspectSkill
- chaos-testSkill
- component-buildSkill
- gofastr-docsSkill
- gofastr-hostSkill
- gofastr-mcp-debugSkill
- gofastr-uiSkill
- log-debugSkill
- plain-wordsSkill
- primitives-sweepSkill
- review-gateSkill
- twin-sweepSkill
The file
We split this skill into the rules it gives an agent
3 rules. 0 are common, 3 are distinctive, and 0 are very specific to this project.
0
Common
Most of the 4 edited versions we read keep these.
3
Distinctive
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 40
If the evidence is thin, say so in one clause in part 1 or part 3.
- Distinctive: line 42
No file paths unless a next step needs one.
- Distinctive: line 43
If the reader asks for detail, give it then, on the one thing asked.
“Common” is measured: kept by at least 6 in 10 of the 4 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Eval report
The reader wants a decision, not a lab notebook. Write these four parts, in this order, with these headings. Nothing before them, nothing after.
1. Is the eval working?
One line: Yes, Partly or No, then one sentence saying why. This is about the eval itself, not the thing it tests. "Partly: the grader matched on wording, so one correct fix scored zero." If the answer is No, stop here and say what has to be fixed first.
2. What is working, and how much
At most four lines, or one small table. Each line is a claim with its number. Always give the fraction with its total, never a bare percentage: "fixed 13 of 30", not "43%". Say what is not working in the same list, with the same kind of number.
3. What I think
Two or three sentences, in the first person. Say whether the direction is right, and the one thing that would change your mind. Take a side.
4. Next steps
At most three, each one line, each starting with a verb. Put the one you would do first at the top. If a step needs the reader's decision, say so on that line.
Rules
- Under 150 words in total, not counting a table.
- Plain words. No method, no list of what was run, no caveats section. If the evidence is thin, say so in one clause in part 1 or part 3.
- Cost goes in part 1 as a clause ("30 runs, $3"), nowhere else.
- No file paths unless a next step needs one.
- If the reader asks for detail, give it then, on the one thing asked.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

