sofer/evaluate
skills/evaluate/SKILL.md
Define evaluation criteria, tests, or quality checks for a deliverable, implementation, workflow, or recurring agent task. Use when defining done, improving a test harness, or checking whether output meets its specification.
$ install --global
skillsauthnpx skillsauth add sofer/.agents evaluateInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
Security Scan Results
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
3
Scanners Passed
9
Scanners in report
Clean
TrivyContainer and dependency vulnerability scanner
95%
Clean
SemgrepStatic code analysis for vulnerabilities
95%
Clean
mcp-scan (Snyk)Model Context Protocol security validation
95%
Skipped
Snyk (dep)Open source security scanning
50%
Skipped
Socket.devSupply chain security analysis
50%
Skipped
VirusTotalMulti-engine malware detection
50%
Skipped
CrowdStrikeAdvanced threat intelligence
50%
Skipped
OSV-ScannerOpen Source Vulnerability database check
50%
Skipped
OWASP Dep-Check
50%
Last scanned: May 1, 2026, 6:42 AM288.5s1 file scanned
SKILL.md
- name:
- evaluate
- description:
- Define evaluation criteria, tests, or quality checks for a deliverable, implementation, workflow, or recurring agent task. Use when defining done, improving a test harness, or checking whether output meets its specification.
Evaluate
Purpose
Make "done" testable before or after work is performed.
Process
- Identify the deliverable or behaviour being evaluated.
- Extract claims about what must be true.
- Convert each claim into a check:
- automated check
- manual check
- human judgement rubric
- Mark each check as:
blocking: failure means not doneimportant: should be fixed, but may not blocknice to have: useful but optional
- Add edge cases most likely to reveal failure.
- For recurring work, define a small reusable eval harness and baseline examples.
Output
## Evaluation checklist
### Automated checks
- [ ] [Check] (blocking|important|nice to have)
- How to verify: [command, test, script, query, or assertion]
### Manual checks
- [ ] [Check] (blocking|important|nice to have)
- How to verify: [steps and expected result]
### Human review criteria
- [ ] [Criterion] (blocking|important|nice to have)
- Pass: [Observable qualities]
- Fail: [Observable failure]
### Edge cases
- [ ] [Scenario]
- Expected behaviour: [Expected result]
Platform mode
For ~/code/fac-cra/, prioritise meaningful verification over speed:
- relevant unit tests
- typecheck and lint
- relevant app build
- Playwright or end-to-end smoke tests
- browser checks for UI
- fixtures, mocks, or harness improvements when current tests are weak
If meaningful verification is not feasible and the change is risky, stop and ask rather than calling the work done.
Related Skills
sofer/agent-config-audit
tools
VerifiedTrustedCommunity
Check whether Claude and Codex have equivalent access to shared agent resources, skills, hooks, plugins, MCP servers, permissions, startup behaviour, and provider-specific adapter config. Use when comparing agent environments, debugging missing capabilities after restart, or deciding whether to symlink a resource or configure a runtime.
4SKILL.mdUpdated May 10, 2026
sofer/skill-usage
testing
VerifiedTrustedCommunity
Record substantive skill use in an append-only local log. Use after choosing or invoking a non-system skill for real work, when a skill is inspected but not used, or when a skill fails to apply. Do not use for routine system skills or incidental file reads.
4SKILL.mdUpdated May 1, 2026
sofer/problem-statement
testing
VerifiedTrustedCommunity
Turn a vague or underspecified request into a self-contained problem statement. Use when the user has a rough idea, when a request would fail if handed directly to an agent, or before non-trivial work that needs shared understanding.
4SKILL.mdUpdated May 1, 2026
sofer/learning
data-ai
VerifiedTrustedCommunity
Append a one-line learning to ~/.agents/learning-log.md. Use when the user types /learning, or when something genuinely worth remembering surfaced during work and the user confirms it should be captured.
4SKILL.mdUpdated May 1, 2026