fusion-issue-authoring
Classify issue type, activate the matching agent mode for type-specific drafting, and enforce shared safety gates before GitHub mutation.
43
43%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./.agents/skills/fusion-issue-authoring/SKILL.mdQuality
Discovery
17%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 attempts to outline an issue-classification and response-drafting workflow for GitHub but relies heavily on internal process jargon ('agent mode,' 'safety gates,' 'GitHub mutation') rather than user-facing language. It lacks a 'Use when...' clause and natural trigger terms, making it difficult for Claude to know when to select this skill over others.
Suggestions
Add an explicit 'Use when...' clause with natural trigger terms like 'GitHub issue,' 'bug report,' 'feature request,' 'triage,' or 'label issues.'
Replace internal jargon ('agent mode,' 'safety gates,' 'GitHub mutation') with concrete user-facing actions such as 'triage incoming GitHub issues,' 'draft responses,' 'apply labels,' or 'create/update issues.'
Include common file or context references users might mention, such as 'issue templates,' '.github/ISSUE_TEMPLATE,' or specific issue types like 'bug,' 'enhancement,' 'question.'
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | It names a domain (GitHub issue handling) and describes some actions (classify issue type, activate agent mode, enforce safety gates), but the actions are somewhat abstract—'activate the matching agent mode' and 'enforce shared safety gates' are internal process descriptions rather than concrete user-facing capabilities. | 2 / 3 |
Completeness | While there is a partial 'what' (classify issues, draft responses, enforce safety), there is no 'when' clause or explicit trigger guidance telling Claude when to select this skill. The absence of a 'Use when...' clause caps this at 2, and the 'what' is also weak enough to warrant a 1. | 1 / 3 |
Trigger Term Quality | The description uses technical jargon like 'agent mode,' 'safety gates,' and 'GitHub mutation' that users would not naturally say. It lacks common trigger terms a user might use such as 'issue,' 'bug report,' 'feature request,' 'triage,' or 'label.' | 1 / 3 |
Distinctiveness Conflict Risk | The mention of GitHub and issue classification provides some specificity, but 'GitHub mutation' and 'safety gates' are vague enough that this could overlap with other GitHub-related skills (e.g., PR review, GitHub Actions). It's somewhat distinguishable but not clearly carved out. | 2 / 3 |
Total | 6 / 12 Passed |
Implementation
70%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a well-structured orchestration skill with excellent workflow clarity and progressive disclosure. Its main weaknesses are moderate verbosity in cache/rate-limit policy sections and a lack of concrete executable examples (e.g., example MCP tool invocations or draft output snippets) that would make the guidance more immediately actionable.
Suggestions
Add at least one concrete example of an MCP tool invocation (e.g., `mcp_github::issue_write` with a sample payload) to improve actionability.
Condense the shared gate cache policy and rate-limit behavior sections into tighter bullet points or move detailed caching logic to a reference file, as much of this is implementation detail that inflates the main skill.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is fairly detailed and well-structured, but includes some verbose sections—particularly the shared gate cache policy and rate-limit behavior blocks—that could be tightened significantly. Some instructions (e.g., explaining what sub-issues are for, what issue-closing keywords do) border on things Claude already knows. | 2 / 3 |
Actionability | The skill provides a clear multi-step workflow with specific MCP tool names and file paths, but lacks any concrete executable examples—no example MCP calls, no example draft output, no example payloads. Guidance is specific in intent but remains descriptive rather than copy-paste ready. | 2 / 3 |
Workflow Clarity | The 8-step workflow is clearly sequenced with explicit validation checkpoints: duplicate check before drafting, draft review before mutation, explicit publish confirmation gate, ordered mutation steps, and relationship validation before linking. Error recovery paths (rate limits, missing MCP server, mutation failures) are well-defined with feedback loops. | 3 / 3 |
Progressive Disclosure | The skill serves as a clear top-level router with well-signaled one-level-deep references to agent files, reference docs, asset templates, and fallback paths. Content is appropriately split between the overview skill and referenced files (agents/, references/, assets/), with clear navigation signals throughout. | 3 / 3 |
Total | 10 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
metadata_field | 'metadata' should map string keys to string values | Warning |
Total | 10 / 11 Passed | |
- Repository
- equinor/fusion-framework
- Commit
060e3af
- 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.