shadow-discovery
Skill1 starsChanged yesterdayMITView on GitHub
Requirements Engineering & Discovery. Use when eliciting software requirements, mapping stakeholders, writing PRD/SRS/BRD, defining FR/NFR, Given/When/Then acceptance criteria, or detecting requirement gaps. Do NOT use for tender KAK pricing (/shadow-presales) or CR disputes (/shadow-pmo).
In this repository51 other files it gives its agents.
- antislop-codeSkill
- antislop-copywritingSkill
- antislop-humanSkill
- antislop-layoutmobileSkill
- antislopSkill
- antislop-uiSkill
- context7Skill
- handoffSkill
- shadow-accessibilitySkill
- shadow-apiSkill
- shadow-architectSkill
- shadow-authSkill
- shadow-bootstrapSkill
- shadow-browserSkill
- shadow-cacheSkill
- shadow-compatSkill
- shadow-concurrencySkill
- shadow-dataSkill
- shadow-dbSkill
- shadow-debugSkill
- shadow-designSkill
- shadow-docSkill
- shadow-domainSkill
- shadow-eaSkill
- shadow-evalSkill
- shadow-finopsSkill
- shadow-gitSkill
- shadow-grcSkill
- shadow-handoverSkill
- shadow-integrateSkill
- shadow-maintainSkill
- shadow-mobileSkill
- shadow-modernizeSkill
- shadow-monarchSkill
- shadow-observeSkill
- shadow-opsSkill
- shadow-perfSkill
- shadow-planSkill
- shadow-pmoSkill
- shadow-presalesSkill
- shadow-privacySkill
- shadow-qualitySkill
- shadow-realtimeSkill
- shadow-refactorSkill
- shadow-releaseSkill
- shadow-resilienceSkill
- shadow-searchSkill
- shadow-secSkill
- shadow-supplychainSkill
- shadow-testSkill
- shadow-uatSkill
The file
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 1 is distinctive, and 4 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 8
Use this skill at the start of any new project, client brief, TOR/KAK analysis, or major feature request to transform raw, ambiguous client …
- Very specific: line 15
Mandatory Evidence Classification: Every discovery claim must be tagged as [EVIDENCE] (grounded in repo code, tickets, analytics, or dated …
- Very specific: line 22
Before proceeding to /shadow-architect or coding, inspect the raw input for:
- Very specific: line 43
When /shadow-discovery is executed, produce or update the following structured sections (compatible with /shadow-doc for .docx generation):
- Very specific: line 52
Every Functional Requirement (FR-MOD-001) MUST have:
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Shadow Discovery (/shadow-discovery) — Requirement Engineering & Scope Gate (SDLC Stage 1)
Use this skill at the start of any new project, client brief, TOR/KAK analysis, or major feature request to transform raw, ambiguous client desires into a contradiction-free, auditable, and testable specification before writing code.
1. The Ambiguity Stop Rule & Evidence Classification (Quality Gate 1)
- Never guess business rules that affect money, legal compliance, access control, or destructive state transitions.
- Mandatory Evidence Classification: Every discovery claim must be tagged as
[EVIDENCE](grounded in repo code, tickets, analytics, or dated interviews),[ASSUMPTION](with risk & validation plan),[PROPOSAL], or[NEEDS VALIDATION]. Never invent fake user research or personas. - The Five Foundational Strategy Questions (
ghaida-intent/skills/strategize/SKILL.md):- Problem Validation: Is this a real, recurring problem worth solving?
- Audience Definition: Who experiences it most acutely, and in what context?
- Solution Fit: Why does this mechanism solve the root cause rather than a symptom?
- Feature Validation: What is the smallest testable slice (MVP) that proves value?
- Competitive & Existing System Landscape: How is it solved today in the codebase or market?
- Before proceeding to
/shadow-architector coding, inspect the raw input for:- Missing Actors / Roles: Who performs the action, who approves it, and who can view/audit it?
- Contradictory Rules: Does Requirement A conflict with Requirement B or existing database invariants?
- Undefined Edge Cases: What happens on cancellation, refund, partial fulfillment, timeout, rejected approval, or duplicate submission?
- Unbounded Scope ("Gold-Plating"): Separate In-Scope (MVP / Phase 1) from Out-of-Scope (Future Phase) explicitly.
1b. Foreign-Key Reference Routing (PM Discovery, Strategy & Research Packs)
Load only the specific reference file needed for deep discovery or strategic framing:
- Evidence & JTBD Framing:
shadow-design/references/ai-product-designer-skill-pack/01-discovery-and-evidence.md - Assumption Identification & Prioritization (
Value | Usability | Viability | Feasibility | Ethics):identify-assumptions-existing/SKILL.md,identify-assumptions-new/SKILL.md&prioritize-assumptions/SKILL.md - Opportunity Solution Tree & Experiment Design (XYZ Hypothesis / Pretotyping):
opportunity-solution-tree/SKILL.md,brainstorm-experiments-existing/SKILL.md&brainstorm-experiments-new/SKILL.md - PRD Authoring & Red-Team PRD Stress Test:
pm-execution/skills/create-prd/SKILL.md&pm-execution/commands/red-team-prd.md - Primary User Research & Synthesis:
ghaida-intent/skills/investigate/SKILL.md&designer-skills/design-research/README.md
2. Mandatory Discovery Deliverables
When /shadow-discovery is executed, produce or update the following structured sections (compatible with /shadow-doc for .docx generation):
A. Stakeholder & Actor Matrix
| Actor / Role | Core Responsibilities | Access Level & Tenant Scope | Key Pain Points / Goals |
Super Admin | Global configuration, tenant provisioning | Cross-tenant (System) | Auditability & zero manual DB edits |
Client Operator | Daily transaction entry & verification | Single-tenant (tenant_id) | Fast entry, clear validation errors |
B. Functional Requirements (FR) with Testable Acceptance Criteria
Every Functional Requirement (FR-MOD-001) MUST have:
- Priority:
P0 (Must-Have)|P1 (Should-Have)|P2 (Nice-to-Have) - Business Rule / Invariant: Explicit formula, state transition (
DRAFT -> SUBMITTED -> APPROVED | REJECTED), and validation constraints. - Given / When / Then Acceptance Criteria: Directly traceable to
/shadow-uatand/shadow-test.
C. Non-Functional Requirements (NFR)
Explicitly quantify:
- Performance & Scale: Target concurrent users, peak RPS, max acceptable
p95latency (< 400ms), and expected annual data volume growth. - Security & Compliance: Authentication method, RBAC/ABAC granularity, PII encryption (UU PDP / ISO 27001 / OWASP ASVS level), and audit log retention.
- Availability & Recovery: Target SLA (
99.9%), RPO (max tolerable data loss window, e.g.< 15 min), and RTO (max recovery time, e.g.< 1 hour).
D. Assumption Risk Matrix & Clarifying Questions for Stakeholder
Map unverified assumptions across 5 Risk Dimensions (Value, Usability, Viability, Feasibility, Ethics), and list open ambiguities in a prioritized table with a Recommended Default so the Monarch or client can approve or override with minimum friction.
3. Stage 1 Exit Gate Checklist (To Unlock /shadow-architect)
- In-Scope vs Out-of-Scope boundaries explicitly documented.
- Every product/UX claim tagged (
[EVIDENCE],[ASSUMPTION],[PROPOSAL],[NEEDS VALIDATION]). - High-risk assumptions (Value, Usability, Viability, Feasibility, Ethics) identified with validation plans.
- All P0 requirements have deterministic Given/When/Then acceptance criteria.
- Zero unresolved contradictions between business rules (Red-Team PRD check passed).
- NFRs (concurrency, latency, security, RPO/RTO) quantified.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

