CtrlK
BlogDocsLog inGet started
Tessl Logo

error-recovery

Standard recovery patterns for all squad agents. When something fails, adapt — don't just report the failure.

56

Quality

65%

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

Fix and improve this skill with Tessl

tessl review fix ./.copilot/skills/error-recovery/SKILL.md

The canonical home for this skill is error-recovery in FritzAndFriends/BlazorWebFormsComponents

SKILL.md
Quality
Evals
Security

Quality

Content

86%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 body is a tight, well-structured instruction skill: every pattern carries concrete parameters, worked examples, and explicit retry/escalation limits, with a useful selection table mapping failure types to patterns. Its only weaknesses are a few abstract judgment steps (alternative selection, criticality assessment) and validation steps confined to the diagnose-and-fix pattern.

DimensionReasoningScore

Conciseness

The body is lean and assumes Claude's competence: each pattern is a terse When/Pattern/Example triple with concrete parameters ("start at 2s, double each attempt", "Maximum 3 retries") and no explanation of concepts Claude already knows. Only the intro paragraph duplicating the frontmatter description is arguably trimmable, which keeps it at level 5 rather than revealing the padding characteristic of level 4.

5 / 5

Actionability

Guidance is mostly executable: numbered steps, concrete parameters (2s/4s/8s backoff, 3-cycle caps), a worked 429 example, and a failure-type selection table. A few steps remain abstract — "identify alternative tool/method" and "Determine if the failed step is critical" give no heuristics for the judgment — which are minor gaps keeping it below fully copy-paste-ready level 5.

4 / 5

Workflow Clarity

Each pattern is a clearly sequenced numbered workflow with explicit feedback loops ("Re-run to verify the fix", "Maximum 3 fix-retry cycles before escalating") and a selection-guide table acting as a checklist. Verification checkpoints appear only within pattern 3 rather than across all workflows, so it sits at level 4 (most checkpoints present) instead of 5.

4 / 5

Progressive Disclosure

This is a single-file skill with no bundle directories, and nothing that belongs in a separate file is inlined: five well-organized pattern sections plus an application section and selection table, all easy to navigate. Per the simple-skill guidance, well-organized sections with no need for external references warrant level 5.

5 / 5

Total

18

/

20

Passed

Description

45%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 names its domain and audience but omits the concrete recovery actions the skill actually contains (retry with backoff, fallbacks, diagnose-and-fix, escalation, graceful degradation), and lacks an explicit "Use when..." invocation trigger. Its trigger condition is generic enough to risk overlap with general error-handling skills.

Suggestions

Add an explicit "Use when..." clause naming invocation triggers, e.g., "Use when a squad agent hits a failure: an API error, tool failure, build/test failure, or exhausted retries."

Enumerate the concrete recovery actions in the description — retry with backoff, fallback alternatives, diagnose-and-fix, escalate with context, graceful degradation — instead of the generic "adapt".

Include natural trigger keywords users would actually say ("retry", "error handling", "fallback", "backoff") to improve both trigger-term quality and distinctiveness from generic troubleshooting skills.

DimensionReasoningScore

Specificity

"Standard recovery patterns for all squad agents" names the domain but lists no concrete actions; "adapt — don't just report the failure" is an exhortation rather than a capability. This matches the anchor for naming the domain with minimal/generic actions, and falls short of level 3 because no concrete actions (e.g., retry with backoff, fallback alternatives) are stated.

2 / 5

Completeness

The "what" is clear (standard recovery patterns for squad agents), but "When something fails, adapt" reads as behavioral instruction to the agent rather than an explicit "Use when..." invocation trigger, so the when-clause is only weakly serving as trigger guidance. Per the guideline capping completeness at 3 for missing or weak explicit trigger guidance, this sits at level 3 rather than 4.

3 / 5

Trigger Term Quality

Terms like "recovery", "fails", and "failure" are relevant, but common natural variations users would say — "retry", "error handling", "fallback", "backoff" — are absent. Some relevant keywords exist, but coverage of synonyms and variations is missing, matching the middle anchor.

3 / 5

Distinctiveness Conflict Risk

"for all squad agents" scopes the skill to a specific niche, but "When something fails" is generic and would overlap with any error-handling or troubleshooting skill. It is somewhat specific yet still carries real overlap risk with similar skills, matching the middle anchor rather than the mostly-distinct level 4.

3 / 5

Total

11

/

20

Passed

Validation

93%

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

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

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

Warning

Total

15

/

16

Passed

Repository
microsoft/waza
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.