skills.rest
open-mercato
open-mercatoOFFICIAL
Explore all skills by this author →
WebsiteGitHub

om-gap-analysis

Verifies client requirements against a platform codebase and produces a gated Epic/Story coverage report.

1.7k|382|Updated Sep 10, 2025
Search Tags:#gap analysis# backlog#Requirements#codebase verification#coverage report#engagement scoping#epic story tree
One-click install
npx skills add https://github.com/open-mercato/open-mercato --skill om-gap-analysis
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: om-gap-analysis
Source: https://github.com/open-mercato/open-mercato/tree/main/.ai/skills/om-gap-analysis
Command: npx skills add https://github.com/open-mercato/open-mercato --skill om-gap-analysis

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Scoping a client engagement against an existing platform usually means weeks of manual code reading and guesswork about what the platform already covers. This Skill turns a folder of client documents (transcripts, requirement dumps, spec docs) into an evidence-backed Epic/Story tree where every coverage verdict is re-verified by executable gates against a validated checkout of the platform's code, so no claim enters the report unverified.

Core Features & Use Cases

  • Three-phase pipeline: Scoping (docs → structured Epic/Story tree), Verification (parallel read-only subagents investigate the platform codebase, every finding gated before writing), and Synthesis (client-facing summary plus prioritized backlog).
  • Executable trust gates: five deterministic scripts enforce completeness of the tree, freshness of the platform checkout, re-run of every grounding query, pipeline citation cross-checks, and surfacing of significant open PRs.
  • License-tier tagging and effort scoring: verdicts are split by core vs licensed tiers and gaps are scored in atomic commits (0–5), never T-shirt sizes.
  • Use Case: A consultancy receives 40 requirement documents for a B2B portal. The Skill produces a coverage report showing which stories the platform already implements, which are partial, and a sequenced backlog of the remaining work with effort estimates.

Quick Start

Run the om-gap-analysis skill on my folder of client requirement documents to produce a verified coverage report and implementation backlog against the platform repository.

Frequently Asked Questions about om-gap-analysis

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run a gap analysis of client requirements against an existing platform?▼

Point the skill at a directory of client documents with a project slug; Phase 1 builds an Epic/Story tree, then after clearing context, re-invoke it on the generated markdown file to run verification and synthesis. The output is a coverage summary and a sequenced backlog.

How are coverage verdicts verified against the platform code?▼

Each story is investigated by a read-only subagent that names a decisive search term, and the gap-validate-finding gate re-runs that search against a validated, freshly fetched checkout of the platform before the verdict is written. Subagent claims are never trusted without the gate passing.

Can it analyze a single capability question instead of a full document set?▼

Yes. For one question like whether the platform supports a feature, it answers directly with the same evidence discipline using a validated checkout search, without starting the batch pipeline designed for directories of documents.

What happens if the analysis run is interrupted mid-verification?▼

Each story carries a status field (pending, done, needs-review) in the source markdown file, so re-invoking the skill on that file resumes the run and skips stories already marked done.

What are the limitations of the automated verdict gates?▼

The gates are falsifiers, not proof oracles: a search hit confirms a string match, not that the code satisfies the acceptance criteria, and verdicts grounded on an integration branch may include code not yet in a tagged release. Semantic judgment remains with the investigating subagent.

Related Skills

Explore More Product & Management Skills→
Benmore-StudioBenmore-Studio
1

feature-alignment

Extract features and estimate scope from meeting transcripts and client documents.

Official
Advanced
alirezarezvanialirezarezvani
110

roadmap-generator

Convert architecture designs into phased roadmaps with Epics, Stories, and Tasks.

Community
Advanced
Elevaria-biaElevaria-bia

aiox-qa

Audit story requirements and test strategy to produce QA gate decisions.

Community
Advanced
udecodeudecode
16.5k

grill-me

Interrogates plans and designs through structured questioning until reaching shared understanding.

Official
Basic