CtrlK
BlogDocsLog inGet started
Tessl Logo

continual-learning

Nightly refinement of an existing per-repo review-style prompt using this reviewer's own finding outcomes. Read confirmed (resolved-by-commit / thumbs-up) and dismissed (thumbs-down) findings, promote the bug patterns the team actually fixes, demote the false-positive patterns, reconcile against the current prompt, and save the refined version. Use this once outcomes exist; use bootstrap-repo-analysis for a cold-start repo.

75

Quality

92%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

92%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.

A lean, well-sequenced instruction skill with concrete tool-call guidance and an explicit empty-outcomes guardrail. The only gap is the absence of a concrete before/after prompt example to ground the abstract reconciliation verbs.

Suggestions

Add a short before/after snippet showing one rule being promoted (or one false-positive being added to the 'do not flag' section) to make the reconcile step copy-paste concrete.

Show the expected shape of the analysis_summary string (e.g. a one-line template) so the 'names what changed this cycle' guidance is unambiguous.

Specify the light gh top-up step more concretely (e.g. a sample gh pr list invocation and how many PRs to sample) so 'Optionally do a light gh top-up' is executable rather than aspirational.

DimensionReasoningScore

Conciseness

The body is lean and targeted with no concept-padding Claude already knows; every line ('promote the recurring ones... quoting the file/diff_hunk context', 'Look for repetition, not one-offs') earns its place.

5 / 5

Actionability

Concrete tool calls with named parameters and constraints ('Call save_review_style_prompt once with the refined custom_prompt (400–1200 words)', 'analysis_summary that names what changed') are actionable, but the prompt-editing guidance ('strengthen rules', 'remove or soften rules') stays abstract without a before/after example.

4 / 5

Workflow Clarity

A clearly numbered 1-2-3 sequence (Read outcomes, Reconcile, Save) with an explicit validation/feedback loop: 'If outcomes were empty and nothing changed... re-save the existing prompt unchanged rather than degrading it.'

5 / 5

Progressive Disclosure

Under 50 lines with no need for external references and three well-organized numbered sections, satisfying the simple-skill exception for progressive disclosure.

5 / 5

Total

19

/

20

Passed

Description

92%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.

A strong, third-person description that concretely states multiple actions and provides an explicit use-when trigger with a clear boundary against a sibling skill. It is slightly jargon-heavy in its trigger terms but otherwise comprehensive and distinct.

DimensionReasoningScore

Specificity

Lists multiple concrete actions ('Read confirmed... and dismissed... findings, promote the bug patterns..., demote the false-positive patterns, reconcile against the current prompt, and save the refined version') with comprehensive coverage of the refinement workflow.

5 / 5

Completeness

Explicitly answers both what ('Nightly refinement of an existing per-repo review-style prompt using this reviewer's own finding outcomes') and when ('Use this once outcomes exist; use bootstrap-repo-analysis for a cold-start repo') with concrete trigger phrases.

5 / 5

Trigger Term Quality

Good coverage of relevant terms ('outcomes', 'findings', 'review-style prompt', 'cold-start repo', 'bootstrap-repo-analysis') but the language is somewhat jargon-heavy and lacks a few natural user-facing synonyms.

4 / 5

Distinctiveness Conflict Risk

Clear niche (refining an existing prompt from accrued outcomes) and explicitly distinguishes itself from the sibling cold-start skill ('use bootstrap-repo-analysis for a cold-start repo'), minimizing conflict risk.

5 / 5

Total

19

/

20

Passed

Validation

100%

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

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
langchain-ai/open-swe
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.