deepgram-incident-runbook
Execute Deepgram incident response procedures for production issues. Use when handling Deepgram outages, debugging production failures, or responding to service degradation. Trigger: "deepgram incident", "deepgram outage", "deepgram production issue", "deepgram down", "deepgram emergency", "deepgram 500 errors".
82
80%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./plugins/saas-packs/deepgram-pack/skills/deepgram-incident-runbook/SKILL.mdQuality
Discovery
89%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is a well-structured skill description with strong trigger terms and clear 'what/when' guidance. Its main weakness is the lack of specific concrete actions—'execute incident response procedures' is somewhat vague and could be improved by listing 2-3 specific actions the skill performs. The explicit trigger term list is a notable strength for skill selection.
Suggestions
Replace the abstract 'execute incident response procedures' with specific actions like 'check Deepgram API health endpoints, analyze error logs, identify failing services, and follow escalation runbooks'.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain (Deepgram incident response) and a general action (execute incident response procedures), but doesn't list specific concrete actions like 'check service health endpoints, review error logs, restart services, escalate to on-call engineers'. The phrase 'incident response procedures' is somewhat abstract. | 2 / 3 |
Completeness | Clearly answers both 'what' (execute Deepgram incident response procedures for production issues) and 'when' (handling outages, debugging production failures, responding to service degradation) with explicit trigger terms listed separately. | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural trigger terms users would actually say: 'deepgram incident', 'deepgram outage', 'deepgram down', 'deepgram 500 errors', 'deepgram emergency', 'deepgram production issue'. These are realistic phrases someone would use when experiencing Deepgram production issues. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive due to the specific vendor name 'Deepgram' combined with incident response context. Very unlikely to conflict with other skills unless there are multiple Deepgram-related skills, and even then the incident/outage focus narrows it clearly. | 3 / 3 |
Total | 11 / 12 Passed |
Implementation
70%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a highly actionable and well-structured incident runbook with executable code for every severity level and a clear escalation workflow. Its main weakness is that all content is crammed into a single monolithic file — the detailed TypeScript implementations and post-incident template should be split into referenced files to improve token efficiency and progressive disclosure. The workflow sequencing and validation checkpoints are strong.
Suggestions
Extract the TypeScript fallback service class, SEV2 mitigation code, and post-incident template into separate bundle files (e.g., fallback-service.ts, sev2-mitigation.ts, post-incident-template.md) and reference them from SKILL.md
Trim the SKILL.md to a concise overview with the triage script inline and severity classification table, linking to detailed response procedures in separate files
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is fairly long (~200+ lines) with substantial code blocks. While most content is actionable, some sections are verbose — the full fallback service class with replay queue is extensive, and the post-incident template could be referenced externally. The severity table and quick reference are efficient, but overall the skill pushes token budget limits. | 2 / 3 |
Actionability | Excellent actionability throughout: the bash triage script is fully executable and copy-paste ready, the TypeScript fallback service is complete and functional, the SEV2 diagnostic code tests models and features systematically, and the mitigation configs are concrete. Every severity level has specific, executable guidance. | 3 / 3 |
Workflow Clarity | The workflow is clearly sequenced from triage (Step 1) through severity-specific responses (Steps 2-4) to post-incident review (Step 5) and escalation (Step 6). Validation is built into the triage script, the SEV2 response includes automatic escalation logic ('All models failing — escalate to SEV1'), and the fallback service includes a replay mechanism with error handling for recovery. | 3 / 3 |
Progressive Disclosure | All content is monolithically packed into a single file with no bundle files or external references for detailed content. The full fallback service implementation, post-incident template, and all severity response code are inline. The post-incident template and the detailed TypeScript classes would be better as separate referenced files, keeping SKILL.md as a concise overview with pointers. | 1 / 3 |
Total | 9 / 12 Passed |
Validation
81%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 9 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
allowed_tools_field | 'allowed-tools' contains unusual tool name(s) | Warning |
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 9 / 11 Passed | |
- Commit
3a2d27d
- Reviewed
Table of Contents
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.