CtrlK
BlogDocsLog inGet started
Tessl Logo

darwinian-evolver

Evolve prompts/regex/SQL/code with Imbue's evolution loop.

57

Quality

67%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./optional-skills/research/darwinian-evolver/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

81%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The skill body is a strong, execution-first document: concrete commands, precise API contracts, a genuinely useful hyperparameter table, and pitfalls that encode real operational failure modes. Its main defects are a missing referenced template file (custom_problem_template.py) and a few prose sections that could be tightened.

Suggestions

Ship the missing templates/custom_problem_template.py referenced in 'Defining a Custom Problem', or inline a minimal complete template since the bundle currently has no templates/ directory.

Trim the 'Status: thin wrapper' paragraph and condense the license note to a single directive line to save context tokens.

Move the external reading links (arXiv, Imbue posts) into a references/ file so the SKILL.md body stays purely operational.

DimensionReasoningScore

Conciseness

The body is dense and nearly all of it is non-obvious, upstream-specific knowledge (install layout, hyperparameter table, nested-pickle snapshots, hardcoded Anthropic CLI) with no padding explaining concepts Claude already knows. Minor trimming opportunities remain — the 'Status: thin wrapper' paragraph and parts of the license note — so it fits 'Efficient; minor instances that could be trimmed' rather than the 5-anchor's every-token-earns-its-place.

4 / 5

Actionability

Install, verify, both quick-starts, snapshot inspection, and the final verification block are all copy-paste-ready commands, and the Organism/Evaluator/Mutator contracts are given with exact signatures. It falls short of 5 because the referenced 'templates/custom_problem_template.py' is missing from the bundle and only the prompt case has a complete worked example, leaving regex/SQL/code without executable coverage.

4 / 5

Workflow Clarity

The sequence is explicit and validated end-to-end: install → 'Verify:' block → smoke-test run → expected outputs stated → custom problem → verification script with exit code 0, plus error-recovery guidance in the pitfalls (try/except returning '<LLM_ERROR: ...>', 'Return [] on parse failure'). This matches 'Clear sequence with explicit validation steps; feedback loops for error recovery'.

5 / 5

Progressive Disclosure

The body is well-sectioned with real one-level-deep bundle references ('scripts/parrot_openrouter.py', 'scripts/show_snapshot.py' both exist) and clean navigation. It is not 5 because the body cites 'templates/custom_problem_template.py', which does not exist in the bundle — a broken reference — leaving 'minor organization gaps' per the 4-anchor.

4 / 5

Total

17

/

20

Passed

Description

53%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is concise and specific about the domain and mechanism, but it omits any 'when to use' trigger guidance and lacks the natural synonyms ('optimize', 'improve') a user would most likely say. It is a clear, non-conflicting description that stops just short of being a complete trigger surface.

Suggestions

Add an explicit 'when' clause, e.g. 'Use when the user asks to optimize or auto-improve a prompt, regex, SQL query, or code snippet against a measurable scorer.'

Include natural trigger synonyms such as 'optimize', 'improve', and 'auto-refine' alongside 'evolve', since users rarely say 'evolve my prompt'.

Optionally name the fitness-function prerequisite in the description to sharpen the boundary against gradient-based optimizers (DSPy, etc.).

DimensionReasoningScore

Specificity

The description names a concrete domain ("prompts/regex/SQL/code") and one action ("Evolve") plus the mechanism ("Imbue's evolution loop"), but offers only a single verb rather than the several distinct actions of the 4-anchor ('Extracts text, fills forms, converts pages'). It is not the generic vagueness of 1–2, but coverage is not comprehensive.

3 / 5

Completeness

The 'what' is clear ('Evolve prompts/regex/SQL/code with Imbue's evolution loop'), but there is no 'Use when...' clause or equivalent explicit trigger guidance, which caps completeness at 3 per the judging guidelines. It is not 4 because the 'when' is entirely absent rather than merely weak, and not 2 because the 'what' is specific, not vague.

3 / 5

Trigger Term Quality

Relevant keywords are present ("evolve", "prompts", "regex", "SQL", "code", "evolution loop"), but common variations users would naturally say — "optimize this prompt", "improve this regex", "auto-improve" — are missing, matching the 3-anchor 'Some relevant keywords but missing common variations or synonyms'. It is above 2 because the terms are domain-specific rather than generic, but below 4 because the most natural phrasing ('optimize') is absent.

3 / 5

Distinctiveness Conflict Risk

The niche is clear — 'Imbue's evolution loop' is a distinctive trigger and 'evolve prompts/regex/SQL' is unlikely to fire for unrelated skills — with only minor overlap risk with generic code-optimization or prompt-tuning skills due to the broad word 'code'. This matches 'Mostly distinct; minor overlap risk' rather than 5's 'minimal conflict risk'.

4 / 5

Total

13

/

20

Passed

Validation

81%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 13 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

metadata_field

'metadata' should map string keys to string values

Warning

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

13

/

16

Passed

Repository
NousResearch/hermes-agent
Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.