agentleFS
jhonatan-oliveiradevjhonatan-oliveiradev/agent-skills/skills/assessing-developer-proficiency/SKILL.md

assessing-developer-proficiency

Skill1 starsChanged 3 months agoMITView on GitHub

Use when a developer needs an evidence-aware diagnosis of current capabilities, confidence, unknowns, and weakly demonstrated areas before deciding what to improve.

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 5 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    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
  1. Distinctive:

    Start with the target role or capability set, then collect only evidence relevant to the current-state question:

    line 11
  2. Distinctive:

    Keep claims separate from observations.

    line 18
  3. Distinctive:

    Use the capability definitions supplied by the product, assessment blueprint, or explicit evaluation contract.

    line 21
  4. Distinctive:

    Do not invent a new definition of proficiency because the available evidence is inconvenient.

    line 28
  5. Distinctive:

    Prefer named capability gaps over a synthetic overall score.

    line 44

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Assessing Developer Proficiency

Establish a current capability baseline from evidence, not confidence theater. The result should make demonstrated strengths, uncertainty, and missing evidence explicit enough for another method to act on.

Build the evidence inventory

Start with the target role or capability set, then collect only evidence relevant to the current-state question:

  • self-reported experience and confidence;
  • conceptual explanations or knowledge checks;
  • code-reading and technical-reasoning observations;
  • implementation or debugging performance;
  • authentic project artifacts and their provenance.

Keep claims separate from observations. Missing evidence is an unknown, not a low score.

Map evidence to fixed criteria

Use the capability definitions supplied by the product, assessment blueprint, or explicit evaluation contract. For each capability:

  1. identify which observable criteria are supported;
  2. respect required evidence dimensions and blocking gates;
  3. determine the highest level actually demonstrated;
  4. record the evidence IDs or artifacts supporting that conclusion;
  5. retain contradictions rather than averaging them away.

Do not invent a new definition of proficiency because the available evidence is inconvenient.

Derive confidence separately

Describe confidence from evidence strength, diversity, consistency, quantity, recency, and provenance. A strong level with old or contradictory evidence may legitimately be proficient / low confidence.

Self-report alone cannot establish proficient or advanced capability. External unverified artifacts may inform the baseline, but should remain explicitly unverified until independently reproduced or checked.

Report the baseline

Return a compact diagnosis with:

  • demonstrated capability state;
  • confidence for each state;
  • evidence used;
  • unknown or weakly evidenced areas;
  • contradictions or stale evidence;
  • recommended reassessment targets where uncertainty matters.

Prefer named capability gaps over a synthetic overall score.

Ownership boundaries

This method owns current-state diagnosis and uncertainty. It does not sequence a roadmap, teach the missing material, redefine assessment criteria, or prioritize gaps from job-market demand. Hand those decisions to their dedicated methods.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.