requirement-summarizer
Produces a structured summary of a requirements document — the key obligations, grouped by actor and concern, with the MUST/SHOULD/MAY breakdown. Use when onboarding to a large spec, when deciding what to implement first, or when the user asks what a 200-page standard actually requires.
Install with Tessl CLI
npx tessl i github:santosomar/general-secure-coding-agent-skills --skill requirement-summarizer95
Quality
93%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Discovery
100%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 an excellent skill description that clearly articulates specific capabilities (structured summaries with obligation breakdowns by actor), uses natural trigger terms users would actually say ('200-page standard', 'spec', 'what to implement'), and explicitly defines when to use it. The RFC-style MUST/SHOULD/MAY terminology creates a distinctive niche that minimizes conflict with generic summarization skills.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: 'structured summary', 'key obligations', 'grouped by actor and concern', 'MUST/SHOULD/MAY breakdown'. These are precise, actionable outputs. | 3 / 3 |
Completeness | Clearly answers both what ('structured summary of requirements with obligations grouped by actor/concern and MUST/SHOULD/MAY breakdown') AND when ('onboarding to a large spec', 'deciding what to implement first', 'user asks what a standard requires'). | 3 / 3 |
Trigger Term Quality | Includes natural terms users would say: 'requirements document', 'spec', 'onboarding', 'what to implement first', '200-page standard', 'what a standard actually requires'. Good coverage of realistic user language. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive niche: requirements documents with RFC-style obligation levels (MUST/SHOULD/MAY). The specific domain of specs/standards and the structured obligation analysis makes it unlikely to conflict with general document summarization skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
87%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
Excellent skill that demonstrates rather than explains. The worked OAuth example is particularly effective at showing the expected transformation. The density targets and 'Do not' section provide clear guardrails. Minor weakness: the actual summarization workflow could be more explicitly sequenced.
Suggestions
Add an explicit numbered workflow: 1. Read spec end-to-end, 2. Mark normative statements, 3. Group by chosen axis, 4. Verify compression ratio, 5. Check all MUSTs have section refs
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Every section earns its place. No explanation of what requirements documents are or why summarization matters. Tables are dense and scannable. The worked example demonstrates rather than explains. | 3 / 3 |
Actionability | Provides concrete output format template, specific density targets (20:1 compression), exact signal words to look for, and a complete worked example showing input→output transformation. Copy-paste ready structure. | 3 / 3 |
Workflow Clarity | The process is implicit rather than explicit: read spec → identify normative statements → organize by axis → format output. The 'Do not summarize a spec you haven't read end-to-end' hint suggests a workflow but doesn't sequence the steps with validation checkpoints. | 2 / 3 |
Progressive Disclosure | Self-contained skill with appropriate length (~100 lines). Well-organized with clear sections (normative vs informative, organizing, worked example, do not, output format). No need for external references given the scope. | 3 / 3 |
Total | 11 / 12 Passed |
Validation
100%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
- 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.