skill-native-escalation-routing

Use when choosing native or multi-LLM handling for init, review, or security requests

Quality

65%

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/skill-native-escalation-routing/SKILL.md

Quality

Content

71%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 routing policy with concrete criteria and example phrases, scoring well on conciseness and actionability. Its main defects are a dangling reference to a host-adapter file that is absent from the bundle and an escalation path that stops at naming commands without saying what to execute.

Suggestions

Fix the dangling reference: either bundle `skills/blocks/codex-host-adapter.md` with the skill or remove/replace the pointer to it in the host note.

Complete the escalation path: state concretely what to do when escalating (e.g., invoke the named /octo:* command, or verify host subagent availability before delegating).

Trim the duplication between the Core Policy lists and the Suggested User Framing quote, and add an explicit precedence rule for requests that match both the native and escalation criteria.

Dimension	Reasoning	Score
Conciseness	The body is lean and assumes Claude's competence — it lists commands and criteria without explaining concepts — but the Core Policy command lists and the Suggested User Framing quote restate each other, so it is efficient with minor trimming possible (anchor 4, not the every-token-earns-its-place of anchor 5).	4 / 5
Actionability	Concrete guidance throughout: named commands (/init, /octo:review), explicit routing criteria, and example user phrases. It falls short of anchor 5 because it never states what to actually do once escalation is chosen beyond naming the slash commands, and no code/command examples for the escalation path itself.	4 / 5
Workflow Clarity	The decision procedure is well sequenced — native criteria checklist, escalation checklist, then explicit ambiguity handling ("prefer Claude-native first and mention Octopus") — but precedence when both criteria sets partially match is only implied, keeping it below the fully explicit anchor 5.	4 / 5
Progressive Disclosure	The body is short and cleanly sectioned, which alone would support a high score, but it points to `skills/blocks/codex-host-adapter.md`, which does not exist anywhere in the skill (no references/, scripts/, or assets/ directories are present), so the one reference it makes is dangling — matching anchor 3 rather than the well-signaled, valid references of anchor 4.	3 / 5
	Total	15 / 20 Passed

Description

58%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 has an explicit 'Use when' clause and names its routing domain, but it captures only the native side of the routing decision: the escalation trigger phrases users actually say (multiple opinions, adversarial review, provider names) are missing, and it risks triggering on ordinary init/review/security requests.

Suggestions

Add escalation trigger terms users naturally say, e.g. 'multiple model opinions', 'adversarial review', 'second opinion', or explicit provider names, so the description triggers on both sides of the routing decision.

Sharpen the 'what': state the outcome explicitly, e.g. 'Routes init, review, and security requests to Claude-native commands or escalates to multi-LLM Octopus workflows.'

Distinguish the trigger from the underlying skills, e.g. 'Use when it is unclear whether a plain /init, /review, or /security-review suffices or multi-LLM escalation is warranted.'

Dimension	Reasoning	Score
Specificity	"choosing native or multi-LLM handling" names the domain and one concrete action across three task types ("init, review, or security requests"), but it does not list several specific actions, so it sits at anchor 3 rather than the multi-action coverage of anchor 4.	3 / 5
Completeness	It has both a clear what ("choosing native or multi-LLM handling") and an explicit "Use when..." trigger clause, but the when could be more specific about escalation triggers, matching anchor 4 rather than the fully concrete trigger phrases of anchor 5.	4 / 5
Trigger Term Quality	"init, review, or security requests" are natural phrases users would say, but the common escalation variations — "multiple model opinions", "adversarial review", "second opinion", provider names — are absent, matching anchor 3 (missing common variations or synonyms) rather than anchor 4's good coverage.	3 / 5
Distinctiveness Conflict Risk	The routing niche is somewhat specific, but "init, review, or security requests" overlaps heavily with the very skills it routes among (e.g., a plain "review my PR" request would match both this and the review skill itself), fitting anchor 3's 'could still overlap with similar skills'.	3 / 5
	Total	13 / 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: nyldn/claude-octopus
Path: skills/skill-native-escalation-routing/SKILL.md
Commit: 0093dab

Reviewed: 5 days 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.