plano-filter-guardrails

Harden Plano filter chains and guardrails. Use for MCP filter setup, prompt guard responses, and safe filter ordering.

Quality

74%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/plano-filter-guardrails/SKILL.md

Quality

Content

78%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 tight, well-structured instruction-only skill with a clear trigger list, a concrete ordering rule, and a verification step, wasting no tokens. Its weakness is actionability: without any example filter configuration, rejection-message template, or test-prompt samples, Claude must invent the concrete mechanics of each step.

Suggestions

Add a minimal copy-paste example of an MCP filter configuration showing the `type`, `transport`, and `tool` fields referenced in step 1.

Include a short template for a clear, actionable guardrail rejection message (step 2) so the desired output shape is unambiguous.

Add a feedback loop after the verification step, e.g., 'If a blocked prompt passes or an allowed prompt is blocked, adjust the chain order or guard rules and re-verify.'

Dimension	Reasoning	Score
Conciseness	The ~35-line body is lean and efficient with no concept explanations Claude already knows, no padding, and every line carrying a directive ('Order chain as guards -> rewriters -> enrichment -> output checks', 'Prevent expensive enrichment on unsafe requests'). It matches the 'every token earns its place' anchor.	5 / 5
Actionability	Guidance is partly concrete — it names specific fields ('Configure filter `type`, `transport`, and `tool` explicitly for MCP') and an explicit ordering rule — but there is no executable example: no filter config snippet, no sample rejection message, and no test-prompt examples. This fits 'some concrete guidance but incomplete; missing key details' better than level 4, which expects mostly executable commands or code.	3 / 5
Workflow Clarity	The numbered checklist gives a clear 1-5 sequence and ends with an explicit verification step ('Verify with representative blocked and allowed test prompts'). It falls short of level 5 because there is no feedback loop — no instruction on what to do when verification fails — and the earlier steps lack inline checkpoints.	4 / 5
Progressive Disclosure	The skill is under 50 lines, single-purpose, with no bundle files and no need for external references; the body is organized into clear 'When To Use', 'Apply These Rules', and 'Execution Checklist' sections. Per the rubric's simple-skill guideline, this earns a 5 on well-organized sections alone.	5 / 5
	Total	17 / 20 Passed

Description

70%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 concise, third-person description that clearly pairs a what with an explicit use-for clause and is anchored to a specific tool niche. Its main limitation is that the stated actions remain category-level ('setup', 'ordering', 'responses') rather than concrete capabilities, and trigger coverage lacks synonyms.

Suggestions

Replace generic action phrases with concrete capabilities, e.g., 'Harden Plano filter chains and guardrails: configure MCP filters (type, transport, tool), tighten guard rejection messages, and reorder chains so guards run before rewriters and enrichment.'

Broaden trigger coverage with natural synonyms users might say, such as 'safety rules', 'request filtering', or 'guardrail rejections', in the Use-for clause.

Make the when-clause more explicit with concrete user phrasing, e.g., 'Use when the user asks to fix filter ordering, set up MCP filters, or improve guardrail rejection behavior.'

Dimension	Reasoning	Score
Specificity	The description names the domain ('Plano filter chains and guardrails') and a few action-ish phrases ('MCP filter setup', 'prompt guard responses', 'safe filter ordering'), but these are generic verbs rather than the multiple concrete actions of the level-4/5 anchors. It is above level 2 ('names the domain but actions are minimal') because setup, responses, and ordering are at least concrete activity categories.	3 / 5
Completeness	Both parts are present: 'what' ('Harden Plano filter chains and guardrails') and an explicit 'when' clause ('Use for MCP filter setup, prompt guard responses, and safe filter ordering'). The 'when' lists three trigger scenarios but could be more explicit with natural user phrasing, matching the level-4 anchor rather than the fully concrete trigger phrases of level 5.	4 / 5
Trigger Term Quality	'MCP filter setup', 'prompt guard responses', 'safe filter ordering', 'guardrails', and 'filter chain' are terms a user working on filters would plausibly say, giving good keyword coverage. It falls short of level 5 because no synonyms or variations (e.g., 'safety rules', 'request filtering', 'content moderation') are included.	4 / 5
Distinctiveness Conflict Risk	'Plano' and 'MCP filter' carve out a clear niche with distinct triggers, but 'guardrails' and 'filter chains' are broad safety vocabulary that could overlap with closely related safety/prompt-hardening skills, fitting the 'mostly distinct; minor overlap risk' anchor.	4 / 5
	Total	15 / 20 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: katanemo/plano
Path: skills/plano-filter-guardrails/SKILL.md
Commit: 72002a6

Reviewed: 20 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.