CtrlK
BlogDocsLog inGet started
Tessl Logo

agent-reproduce-align

Use after a Codex or Claude Code feature has been implemented in Qwen Code to run the selected reference agent and Qwen Code under the same scenario, capture HTTP and terminal traces, compare request bodies, tool/function schemas, outputs, and iterate until the reproduced behavior is close enough.

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

100%

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

The body is a well-structured, concise workflow with executable commands, an explicit patch-rerun-retry feedback loop, and a clear termination checkpoint, supported by real one-level-deep reference and script files. It avoids padding and assumes Claude's competence throughout.

DimensionReasoningScore

Conciseness

The body is lean and assumes Claude's competence, with no concept-explaining fluff; every line is actionable guidance, rules, or commands Claude would not already know, e.g. 'Compare contracts before wording. Exact prompt text is usually implementation detail.'

3 / 3

Actionability

Provides fully executable, copy-paste-ready commands with real flags and paths, e.g. the full normalize_trace.py invocation, compare_traces.py invocation, and run_pair_capture.sh with REPRO_REFERENCE_AGENT=codex and concrete arguments.

3 / 3

Workflow Clarity

A clear numbered sequence (steps 1-8) with an explicit feedback loop ('Patch Qwen Code, rerun the smallest failing scenario, and repeat') and a termination checkpoint ('Stop when Qwen Code passes the user-visible scenario and the remaining trace differences are documented as intentional'), plus a Done Criteria checklist.

3 / 3

Progressive Disclosure

The body is a concise overview pointing to one-level-deep references that all exist (references/alignment-workflow.md and the three scripts), clearly signaled with 'Read references/alignment-workflow.md before the first comparison pass' and invoked by concrete path.

3 / 3

Total

12

/

12

Passed

Description

85%

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 specific, complete, and distinctive, clearly conveying both the workflow it performs and the post-implementation context in which to use it. Its main weakness is trigger phrasing that reads as an internal procedural cue rather than natural user-requested keywords.

Suggestions

Add natural trigger terms a user would actually say, e.g. 'Use when reproducing or verifying an agent's behavior in Qwen Code, comparing it against Codex or Claude Code'.

Soften the procedural 'after a feature has been implemented' framing so the trigger is recognizable from a user request rather than only from an internal workflow state.

DimensionReasoningScore

Specificity

Lists multiple concrete actions ('run the selected reference agent and Qwen Code under the same scenario', 'capture HTTP and terminal traces', 'compare request bodies, tool/function schemas, outputs', 'iterate until the reproduced behavior is close enough'), matching the comprehensive-action anchor rather than the domain-plus-some-actions anchor.

3 / 3

Completeness

Clearly states both what it does (capture traces, compare request bodies/tool-function schemas/outputs, iterate) and when to use it ('Use after a Codex or Claude Code feature has been implemented in Qwen Code'), with an explicit trigger clause rather than an implied one.

3 / 3

Trigger Term Quality

Relevant keywords are present (Codex, Claude Code, Qwen Code, reproduce) and it opens with 'Use after...', but the trigger is procedural/internal ('after a feature has been implemented in Qwen Code') rather than natural user phrasings, and it misses common variations a user would say.

2 / 3

Distinctiveness Conflict Risk

Occupies a narrow, named niche (agent reproduction/parity for Qwen Code) with triggers tied to specific agent names, making it unlikely to conflict with unrelated skills.

3 / 3

Total

11

/

12

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
QwenLM/qwen-code
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.