agent-introspection-debugging

Structured self-debugging workflow for AI agent failures using capture, diagnosis, contained recovery, and introspection reports.

Quality

—

Does it follow best practices?

Impact

—

No eval scenarios have been run

Securityby

Passed

No known issues

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 lean, actionable, and well-sequenced, with strong checklists and a diagnosis table that guide the agent through a clear recovery loop. As a self-contained instruction-only skill it meets the rubric's high-water marks across all dimensions.

Dimension	Reasoning	Score
Conciseness	Dense tables, templates, and ordered heuristics assume Claude's competence without restating known concepts; every section earns its place.	3 / 3
Actionability	Concrete capture/recovery templates and a diagnosis table with specific patterns, causes, and checks give specific guidance; absence of executable code is acceptable for an instruction-only skill per scoring_notes.	3 / 3
Workflow Clarity	A clearly sequenced four-phase loop with explicit checklists, an ordered recovery heuristic, and an output standard including validation/feedback logic.	3 / 3
Progressive Disclosure	A self-contained single file (no references/ bundle exists) with well-organized sections and no nested references; external ECC skill names are navigation hints, not bundled files.	3 / 3
	Total	12 / 12 Passed

Description

67%

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 and occupies a clear niche, but omits an explicit use-trigger clause and leans on jargon rather than natural user keywords. Adding a "Use when..." sentence would raise completeness and trigger-term quality.

Suggestions

Append a "Use when..." clause naming natural triggers (e.g., "Use when the agent is looping, hitting tool-call limits, or burning tokens without progress").

Swap jargon like "introspection reports" for more natural phrasing users would actually say.

Keep the concrete action list (capture, diagnosis, recovery, report) as the strong "what" anchor.

Dimension	Reasoning	Score
Specificity	"capture, diagnosis, contained recovery, and introspection reports" lists multiple concrete named actions matching the anchor for several specific actions.	3 / 3
Completeness	Clearly states what the skill does but provides no explicit "when to use" trigger clause, so it answers only one of the two required questions.	2 / 3
Trigger Term Quality	Relies on technical jargon ("self-debugging workflow", "introspection") with no natural user-facing trigger terms; missing common variations a user would say.	2 / 3
Distinctiveness Conflict Risk	The narrow niche of "AI agent failures" with a distinct introspection framing makes it unlikely to trigger for the wrong skill.	3 / 3
	Total	10 / 12 Passed

Validation

100%

Warnings & errors only

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: affaan-m/ECC
Commit: 4130457

Reviewed: 1 day ago

Table of Contents

Discovery Implementation Validation

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.