fabric-heal
Self-healing skill that improves signal mapper keyword coverage through iterative problem generation and keyword patching. Use when user says "heal signal mapper", "improve keyword coverage", "generate problem statements", "run healing loop", "patch signal mapper", or asks about "signal mapper gaps". Do NOT use for project architecture or deployment.
63
73%
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 ./.github/skills/fabric-heal/SKILL.mdQuality
Discovery
89%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is a well-structured skill description with strong completeness and distinctiveness. It includes explicit 'Use when' triggers and a 'Do NOT use' exclusion, which is excellent for disambiguation. The main weakness is that the specific actions described use domain-specific jargon that may not be fully concrete to someone unfamiliar with the system.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description names the domain ('signal mapper keyword coverage') and some actions ('iterative problem generation', 'keyword patching'), but the actions are somewhat abstract and domain-specific jargon rather than universally concrete. It's not entirely vague but doesn't list multiple clearly understandable concrete actions. | 2 / 3 |
Completeness | The description clearly answers both 'what' (self-healing skill that improves signal mapper keyword coverage through iterative problem generation and keyword patching) and 'when' (explicit 'Use when' clause with multiple trigger phrases, plus a 'Do NOT use' exclusion clause). | 3 / 3 |
Trigger Term Quality | The description includes a strong set of natural trigger phrases: 'heal signal mapper', 'improve keyword coverage', 'generate problem statements', 'run healing loop', 'patch signal mapper', 'signal mapper gaps'. These are specific phrases a user would actually say when needing this skill. | 3 / 3 |
Distinctiveness Conflict Risk | The description is highly specific to a niche domain ('signal mapper', 'keyword coverage', 'healing loop') and explicitly excludes unrelated use cases ('project architecture or deployment'), making it very unlikely to conflict with other skills. | 3 / 3 |
Total | 11 / 12 Passed |
Implementation
57%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill provides a reasonable overview of a two-mode healing process with good structural organization and clear references. Its main weaknesses are the lack of executable command examples for the CLI tools and missing validation/feedback loops in the healing workflow, which are important for an iterative self-healing process. The orchestrator reporting section reads more like documentation of existing behavior than actionable guidance.
Suggestions
Add concrete CLI command examples for both modes, e.g., `python signal-categories-cli.py add --category 3 --keywords 'term1,term2'` and `python signal-mapper.py --input problem-statements.md`
Add an explicit validation checkpoint after keyword patching: re-run the benchmark, check if coverage improved, and define what to do if it didn't (retry with different terms, escalate, etc.)
Trim or relocate the orchestrator reporting section—it describes what the orchestrator does rather than instructing Claude on what to do, and could be moved to a reference file
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Mostly efficient and avoids explaining concepts Claude already knows. However, the orchestrator reporting section includes details about ASCII bar charts and analytics persistence that feel like documentation of existing behavior rather than actionable instructions for Claude. | 2 / 3 |
Actionability | Provides specific tool names (signal-categories-cli.py, signal-mapper.py), concrete constraints, and a clear output format for problem statements. However, it lacks executable command examples (e.g., exact CLI invocations for the tools), and the heal mode instructions are high-level without showing concrete patching steps or example keyword additions. | 2 / 3 |
Workflow Clarity | The two modes are clearly separated and Mode 2 has a basic sequence (map terms → update registry). However, there are no explicit validation checkpoints—no step to verify that added keywords actually improve coverage before cleanup, and no feedback loop for error recovery when patching fails or coverage doesn't improve. | 2 / 3 |
Progressive Disclosure | The skill is well-structured with clear sections for each mode, a constraints section, and a references section pointing to specific files one level deep. For a skill of this size and complexity, the organization is appropriate and navigation is straightforward. | 3 / 3 |
Total | 9 / 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.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
- Repository
- microsoft/fabric-task-flows
- Commit
b986f41
- 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.