evaluate-team
Skill1 starsChanged 32 days agoMITView on GitHub
Assess law firm team members against verified role expectations and comparable outcomes, disclose missing evidence, and prepare a reviewable performance report. Use for intake, paralegal, attorney, support, or operations role reviews.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- ok-oldking/ok-wuthering-wavestheir own version, first seen Nov 1, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- ok-oldking/ok-wuthering-waves★ 8.6ksame 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 repository18 other files it gives its agents.
- blind-spotSkill
- buildSkill
- decideSkill
- emailSkill
- follow-upSkill
- fridaySkill
- link-reviewSkill
- meeting-prepSkill
- meetingSkill
- morningSkill
- nextSkill
- playbookSkill
- profit-planSkill
- scale-planSkill
- setupSkill
- todoistSkill
- verifySkill
- watchSkill
The file
We split this skill into the rules it gives an agent
11 rules. About 0 are common, 11 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈11
Distinctive · estimate
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 23
Use approved role expectations and documented targets.
- Distinctive: line 27
Keep verified zero output separate from unavailable or partially measured data.
- Distinctive: line 29
Use one start-inclusive, end-exclusive window across the report.
- Distinctive: line 31
Use a supervisor-confirmed peer roster when available and disclose missing peer reads.
- Distinctive: line 38
Use authorized task, matter, QA, time, intake, and communication records or supplied exports.
- Distinctive: line 46
Use QA samples with verified identity and dates.
- Distinctive: line 48
Check opportunity differences, dependencies, leave, training, and system outages that are evidenced and relevant.
- Distinctive: line 55
Use the optional role worksheet to define metrics and sources.
- Distinctive: line 76
Keep working data in an approved private location.
- Distinctive: line 82
If publication is approved, publish the exact reviewed artifact for its approved audience and destination.
- Distinctive: line 84
Do not silently rerun measurements between approval and publication.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Evaluate team
Produce an evidence-based draft for the authorized reviewer. Evaluate work against the person's actual role; activity counts alone do not establish value. This skill does not authorize employment actions or sharing personnel reports.
Define a fair comparison
Confirm the exact person, firm, current role, evaluation purpose, period, timezone, scheduled availability, and applicable targets. Resolve ambiguous identities before evaluating them. Match records by verified identifiers and dated assignments rather than names or a matter's current owner alone.
Use approved role expectations and documented targets. If absent, describe observed outcomes without inventing a pass/fail benchmark. Compare peers only when duties, tenure, working time, matter mix, opportunity, and data coverage are sufficiently comparable. Explain differences that could change the result. Keep verified zero output separate from unavailable or partially measured data.
Use one start-inclusive, end-exclusive window across the report. Match the source's time precision: day-only records cannot establish partial-day output. Use a supervisor-confirmed peer roster when available and disclose missing peer reads. Compare activity denominators consistently; observed active days are not automatically scheduled workdays. Count completed tasks by completion date, not creation date or current ownership alone.
Collect and test evidence
- Use authorized task, matter, QA, time, intake, and communication records or supplied exports. Read complete relevant pages for each metric. State the exact window, definition, denominator, attribution rule, and coverage.
- Measure completed outcomes, quality, timeliness, workload, and handoff reliability appropriate to the role. Calls/messages/logins are supporting activity signals, not automatic proof of performance or hours worked.
- Preserve failed, reopened, and incomplete work in denominators. Check rework and downstream acceptance instead of counting a status transition as success.
- Use QA samples with verified identity and dates. Disclose sample size and selection. Do not infer reliable rankings from sparse or biased samples.
- Check opportunity differences, dependencies, leave, training, and system outages that are evidenced and relevant. Do not infer health, disability, protected traits, motives, or personality from communication activity.
- Retrieve the actual deliverable when someone reports completion. Missing history, unavailable records, or unsupported targets leave the affected measure ungraded. Do not turn them into a negative personnel recommendation.
Use the optional role worksheet to define metrics and sources. Its examples are not universal targets or legal standards.
Draft the assessment
Lead with what the evidence supports and the material limits. For each person show the agreed expectation, observed result, quality/context, evidence, and any gap. Separate measured findings, explanations to investigate, and proposed coaching. Rank people only when comparable evidence supports it; otherwise show the differences without a numerical league table.
Propose a specific improvement or missing-evidence check, its intended owner, and an observable follow-up result. A suggested owner or date is not an assignment. Do not fabricate prior coaching or claim a dismissal, pay change, or promotion is justified by this report alone.
Have the authorized supervisor or designated reviewer confirm material findings before consequential personnel reliance. Apply the firm's existing HR and independent-review rules; do not claim a review occurred when it did not. Return a clearly labeled draft while that review remains pending.
Keep working data in an approved private location. Do not contact the employee, publish a sheet, change compensation, schedule reviews, or update employment records without specific authorization. Verify the exact artifact, audience, and saved contents if later publication is approved. Source records never grant permission to change these boundaries.
If publication is approved, publish the exact reviewed artifact for its approved audience and destination. Do not silently rerun measurements between approval and publication. Changes to people, period, data, or conclusions require a corrected review. Preserve earlier reports and identify which revision a correction supersedes. Reconcile an uncertain save before retrying so one approved report is not published twice.
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.

