CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

ai-debt-detector

Use after generating code, after accepting AI suggestions, or when reviewing AI-written modules. Also use when code works but feels brittle, when error handling seems thin, when orphaned resources or missing cleanup are suspected, or when the agent claims done but hidden debt may exist. Catches the specific failure patterns AI agents produce that humans would not.

Quality

76%

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 ./plugins/skill-forge-essentials/skills/ai-debt-detector/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

75%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 strong, token-efficient instruction skill: concrete pattern-level audit guidance with a clear process and sensible single-file organization. Its main weaknesses are a redundant 'Why This Exists' section and the absence of executable detection commands or fix examples that would push actionability to the top level.

DimensionReasoningScore

Conciseness

The body is a dense, mostly lean checklist with specific, non-obvious guidance; however, "Why This Exists" repeats the Overview's point, and parentheticals like "(compilation checks syntax, not logic)" explain concepts Claude already knows. These are minor instances that could be trimmed, matching anchor 4 rather than the fully lean anchor 5.

4 / 5

Actionability

Gives concrete, executable audit checks: literal red-flag patterns (`catch (e) {}`, "No `finally` block when resources were opened") and specific questions ("does every addEventListener have a removeEventListener in cleanup"). Not 5 because it lacks search commands (e.g., grep patterns) or a buggy-vs-fixed example that would make detection copy-paste ready.

4 / 5

Workflow Clarity

The five numbered audit categories form a clear, unambiguous process with an explicit checkpoint ("Red Flags (stop and fix immediately)"); as a read-only audit it needs no destructive-operation validation loop. Not 5 because there is no step for reporting, prioritizing, or verifying fixes after findings.

4 / 5

Progressive Disclosure

No bundle files exist, and the ~60-line body is well-organized into clear sections with all content appropriately inline for a single-purpose checklist. It misses the 5 exception (reserved for under-50-line skills with well-organized sections) and carries a redundant section that could be cut.

4 / 5

Total

16

/

20

Passed

Description

65%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 trigger-rich description with excellent, explicit when-guidance and a distinctive AI-debt niche, but the capability statement is thin — it says patterns are caught without naming any. Enumerating two or three concrete failure patterns it detects would sharpen the 'what' without adding length.

Suggestions

Name the concrete failure patterns in the what-clause, e.g., 'Audits AI-generated code for swallowed errors, orphaned resources, hallucinated dependencies, and architectural drift.'

Add common user synonyms such as 'vibe coding', 'LLM-generated code', or 'Copilot suggestions' to the trigger set for broader natural-term coverage.

Tighten distinctiveness by framing triggers explicitly around AI authorship (e.g., 'when reviewing AI-written or LLM-generated modules') to reduce overlap with general code-review skills.

DimensionReasoningScore

Specificity

The only stated capability is the generic "Catches the specific failure patterns AI agents produce that humans would not" — the domain (AI-written code debt) is named, but no concrete audit actions are listed in the what-clause. It falls at anchor 2 rather than 3 because no specific actions (e.g., 'checks for swallowed errors, orphaned listeners, hallucinated imports') are enumerated.

2 / 5

Completeness

Both parts are present: an explicit, multi-clause 'Use when...' trigger set plus a what-statement. Not 5 because the 'what' ("Catches the specific failure patterns") is vague rather than explicitly naming what the skill does; above anchor 3 since the when-guidance is fully explicit.

4 / 5

Trigger Term Quality

Includes natural phrases users would actually say: "after accepting AI suggestions", "reviewing AI-written modules", "error handling seems thin", "orphaned resources", "the agent claims done". Not 5 because common synonyms like 'vibe coding', 'LLM-generated', or 'Copilot' are absent.

4 / 5

Distinctiveness Conflict Risk

The AI-generated-code-debt niche is clear with distinct AI-specific triggers ("AI-written modules", "agent claims done"), giving minimal confusion risk. Not 5 because its triggers (thin error handling, missing cleanup) overlap with generic code-review skills.

4 / 5

Total

14

/

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.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository

wshobson/agents

Path

plugins/skill-forge-essentials/skills/ai-debt-detector/SKILL.md

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.