Name: testland/chaos-drill-protocol
Rating: 72.8 (1 reviews)
Author: testland

testland/chaos-drill-protocol

Run protocol and run workflow for a chaos experiment that has already been designed: the four pre-flight gates (non-production target, measured healthy baseline, live observability, a rollback that has actually been exercised), how to pick a conservative blast-radius bound, the sampling cadence and abort criteria fixed in writing before injection, the per-runner inject and abort commands (Chaos Mesh / Litmus / Gremlin / Toxiproxy), the refuse-to-start rules (no blast-radius bound, production context, degraded baseline, offline observability, unexercised rollback), and the recovery-validation step with its tolerance and timeout. Owns execution safety only, not experiment design: the steady-state hypothesis, the fault to inject, and the experiment file come from chaos-experiment-author. Use when an experiment definition exists and a fault is about to be injected into a running system, and the go/no-go gates, abort thresholds, and recovery check still need to be agreed and written down before the fault starts.

Quality

91%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Medium

Suggest reviewing before use

Quality

Content

85%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 highly actionable, well-sequenced safety protocol with strong progressive disclosure and explicit validation checkpoints throughout. The main weakness is conciseness — several rationale passages could be trimmed without sacrificing clarity or safety.

Suggestions

Tighten the 'Why this protocol is strict' section and the psychological rationale under 'Abort criteria are written before injection' to one or two sentences each; the gate itself already carries the lesson.

Consolidate the repeated SRE test-induced-emergency references (currently restated in the intro, Gate 4, the downstream-breach row, and Limitations) into a single pointer after the first mention.

Move the load-arithmetic table or the worked-example recap into references/ if they are reference rather than run-time material, to keep the live protocol lean.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient but carries several justificatory passages — the 'Why this protocol is strict' narrative, the psychological rationale for pre-written thresholds, and repeated retellings of the SRE test-induced emergency — that could be tightened without losing the actionable protocol.	3 / 5
Actionability	Copy-paste per-runner inject/abort commands for Chaos Mesh, Litmus, Gremlin, and Toxiproxy, a concrete drill-contract table, explicit gate pass/fail criteria, a recovery-validation table, and a fully worked example with real numbers cover the common cases.	5 / 5
Workflow Clarity	Four stages are clearly sequenced (pre-flight gates, blast-radius, live monitoring, recovery validation) with explicit pass/fail gates, refuse-to-start rules, and an abort→record→recover feedback loop; validation is thorough for this destructive operation.	5 / 5
Progressive Disclosure	The overview points to three real one-level-deep references (golden-signals.md, worked-example.md, anti-patterns.md), each clearly signaled and described, with bulk detail appropriately split out of SKILL.md.	5 / 5
	Total	18 / 20 Passed

Description

92%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 specific, complete, and well-scoped description that names concrete actions, gives an explicit 'Use when' trigger, and draws clear boundaries against sibling skills. Only mildly lacking in natural-term synonym coverage.

Dimension	Reasoning	Score
Specificity	The description enumerates multiple concrete actions — the four pre-flight gates, conservative blast-radius bound selection, sampling cadence and abort criteria fixed before injection, per-runner inject/abort commands (Chaos Mesh/Litmus/Gremlin/Toxiproxy), refuse-to-start rules, and recovery validation with tolerance and timeout — giving comprehensive coverage.	5 / 5
Completeness	It explicitly answers both what (the run protocol and its enumerated components) and when ('Use when an experiment definition exists and a fault is about to be injected into a running system, and the go/no-go gates, abort thresholds, and recovery check still need to be agreed and written down before the fault starts').	5 / 5
Trigger Term Quality	Strong domain keywords appear ('chaos experiment', 'fault is about to be injected', 'blast-radius', 'abort thresholds', 'recovery check') with a concrete 'Use when' clause, but a few natural synonyms a user might say (e.g. 'chaos test', 'fault injection') are absent.	4 / 5
Distinctiveness Conflict Risk	It carves a clear niche (execution safety only) and explicitly disambiguates from chaos-experiment-author (hypothesis/fault/experiment file) and failure-injection-test-author, minimizing wrong-skill triggering.	5 / 5
	Total	19 / 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.

Reviewed

19 days ago

Table of Contents

Discovery Implementation Validation