ifixai

Guide the user through an independent iFixAi audit of their own agent, checking whether it does the job it is supposed to do given their business rules and org structure. Prefer pointing it at the user's REAL deployed agent over its HTTP endpoint (its actual tools, retrieval, and governance) with `--provider http --endpoint <url>`; only when no endpoint is reachable, fall back to replicating the model beneath as a bare stand-in (Anthropic, OpenAI, Gemini, Azure, Bedrock, etc.). Graded by the judge(s) of their choice (the same model, one independent judge, or a cross-vendor panel). You are the operator who walks them through it and explains the scorecard, running the SAME `ifixai run` engine as the guided CLI. Use when the user asks to run iFixAi or to audit an agent.

Quality

78%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./plugin/skills/ifixai/SKILL.md

Quality

Content

70%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 highly actionable and exceptionally well-sequenced with strong validation checkpoints, but it is markedly over-long and padded with repetition and over-explanation, and it fails progressive disclosure by inlining reference-grade detail and citing a non-existent docs file.

Suggestions

Tighten conciseness: state the http-vs-bare and "never production" guidance once in Step 8 (or a short Containment section) and remove the near-duplicate restatements in Steps 0, 3, 6, and Honest Constraints; cut explanations of things Claude already knows (venv script paths, PowerShell backtick).

Move the evidence-floor table and the full fixture-authoring spec into a real `references/fixture_authoring.md` (create the directory), keeping SKILL.md as an overview that links one level deep; remove or fix the dead `docs/fixture_authoring.md` reference.

Consolidate the provider/key matrix and flag semantics that recur across Steps 0, 6, and 8 into a single reference table so the same env-var and flag detail is not repeated three times.

Dimension	Reasoning	Score
Conciseness	The ~730-line body is noticeably verbose and padded: the http-vs-bare distinction and the "throwaway, never production" warning are restated across Steps 0, 3, 6, 8, and Honest Constraints, and it explains concepts Claude already knows (venv console-script paths, PowerShell backtick continuation). It is above the tutorial-level 1-anchor but sits in the verbose band below the efficient midpoint.	2 / 5
Actionability	It provides copy-paste-ready executable bash commands (e.g. `ifixai run --provider http --endpoint <url> --fixture ifixai-fixture.yaml --grounding sut --mode standard --judge-provider anthropic`), a complete schema-valid fixture YAML, and an env-var key table covering the common cases.	5 / 5
Workflow Clarity	Steps 0–9 form a clearly sequenced workflow with explicit validation checkpoints: a mandatory `--dry-run` estimate, a pre-bill connection test, an explicit "wait for an explicit yes", and per-SUT containment confirmation, plus error-recovery guidance (fix id/key/endpoint and re-run, RUN INVALID handling).	5 / 5
Progressive Disclosure	No bundle files (references/, scripts/, assets/) exist, yet the body inlines ~80 lines of fixture evidence-floor detail and a full YAML spec that belong in a reference file, and its one external reference (`docs/fixture_authoring.md`, section "Evidence floors") points to a file that is absent — structure is present but content is not split out and the reference is dead.	3 / 5
	Total	15 / 20 Passed

Description

87%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 is specific, complete, and distinctive, clearly stating both what the skill does and when to invoke it with concrete trigger phrases. Its only gap is slightly less-than-comprehensive action and trigger-term coverage.

Dimension	Reasoning	Score
Specificity	Names the domain and multiple concrete actions ("Guide the user through an independent iFixAi audit", "checking whether it does the job it is supposed to do given their business rules and org structure", "running the SAME `ifixai run` engine"), but the action enumeration is not fully comprehensive, so it sits between the 3- and 5-anchors at 4.	4 / 5
Completeness	It explicitly answers both what ("Guide the user through an independent iFixAi audit of their own agent…") and when ("Use when the user asks to run iFixAi or to audit an agent."), with concrete trigger phrases, matching the top anchor.	5 / 5
Trigger Term Quality	Natural trigger phrases users would say are present ("run iFixAi", "audit an agent", "agent", "audit") with good coverage, though a few common synonyms like "test my agent" or "evaluate my agent" are missing, placing it just below comprehensive.	4 / 5
Distinctiveness Conflict Risk	It carves a clear niche (iFixAi agent auditing) with distinct triggers ("run iFixAi", "audit an agent") and minimal overlap risk with other skills.	5 / 5
	Total	18 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (736 lines); consider splitting into references/ and linking	Warning

	Total	15 / 16 Passed

Repository: ifixai-ai/iFixAi
Path: plugin/skills/ifixai/SKILL.md
Commit: 643e17e

Reviewed: 3 hours 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.