chaussky/ainalyst/requirements-assess-changes

Claude Code and agent skill by chaussky/ainalyst

Trust tier 2
  • community
LLM review pending
Source check pending

Checking that the SKILL.md file is still in the repository.

BABOK 5.4 skill — Assess Requirements Changes (Change Request). Use this skill when a request to change requirements comes in: you need to assess impact, run a consequence analysis, score the CR, and produce a recommendation (Approve/Defer/Reject). Triggers: "change request", "CR", "request for change", "impact analysis", "assess requirements changes", "assess change", "change to requirements", "new CR".

Install

npx vskill@latest install chaussky/ainalyst/requirements-assess-changes

SKILL.md

SKILL: BABOK 5.4 — Assess Requirements Changes

Task: assess the consequences of a proposed change to requirements and designs (Change Request). MCP server: requirements_assess_changes_mcp.py Reference: references/cr_assessment_guide.md


What this task is about

5.4 is the change gatekeeper. Every CR goes through four steps: open_cr → run_cr_impact → score_cr → resolve_cr

What the BA does in 5.4:

  • Registers the CR and determines the formality level
  • Analyzes impact via the 5.1 traceability graph
  • Scores the CR along five axes (Benefit / Cost / Impact / Schedule / Urgency)
  • Receives a recommendation and hands it off for a decision to the authorized stakeholder
  • Records the decision with rationale, updates requirement statuses

What 5.4 does NOT do:

  • It does not make the final decision — it only provides a recommendation
  • It does not change the content of requirements (that's the BA via 5.2, after Approved)
  • It does not re-prioritize requirements (that's 5.3 — run it after resolve_cr)
  • It does not formally approve requirements (that's 5.5)

Output of 5.4: a CR Decision Record → feeds into 4.4 (communication) and 5.5 (approval/audit).


Inputs from other tasks

TaskWhat it provides
5.1Traceability graph → BFS traversal for impact analysis
5.2Requirements stability → risk flags for volatile requirements
5.3Priorities → check for CR conflicts with Must/Won't
3.3Governance approach → who is authorized to decide on CRs
4.5Decision Log → where the final decision is recorded

When this skill activates

  • A stakeholder requests a new piece of functionality
  • A change in business strategy or scope is identified
  • Legislative/regulatory requirements have changed
  • A developer has identified a technical constraint requiring a requirements review
  • New data has emerged that changes the understanding of the business need
  • Testing has revealed a mismatch between requirements and reality

Four pipeline steps

Step 1 — Open the CR (open_cr)

When: a change request has been received and needs to be registered.

Algorithm:

  1. Determine the type of change: new requirement / change to an existing one / removal / architectural
  2. Identify the initiator and the affected areas
  3. Determine the formality level (read references/cr_assessment_guide.md → "Assessment formality"):
  • Predictive + close to release → high formality
  • Agile + start of iteration → standard
  • Regulatory CR → always high, Urgency = Critical automatically
  1. Call open_cr

Result: the CR is registered as a node in the 5.1 repository, status open.

📌 A CR is a new node type in the 5.1 repository. It is linked to affected requirements
via a modifies relation (added in run_cr_impact).

Step 2 — Impact analysis (run_cr_impact)

When: the CR is open and you need to understand the scope of the change.

Algorithm:

  1. Call run_cr_impact with the CR ID and the list of target requirements
  2. The tool performs a BFS traversal of the 5.1 graph starting from each target requirement
  3. Review the result:
  • List of affected requirements by relation type
  • depends → what will lose meaning
  • verifies → which tests need to be rewritten
  • satisfies → which code needs to be reworked
  • derives → which child requirements are affected
  1. Check: are there any affected requirements with no traceability to a BR?
  2. Check: are there any volatile requirements among those affected (from 5.2)?
📌 If the scope turns out unexpectedly large, notify the initiator before score_cr.
Many CRs get withdrawn from consideration at exactly this step.

Step 3 — Score the CR (score_cr)

When: impact analysis is complete and the real scope is known.

Algorithm:

  1. The technical axes are filled in automatically (Impact + Schedule from step 2)
  2. Enter the business axes manually:
  • Benefit: High / Medium / Low — what does the business gain?
  • Cost: Low / Medium / High — full cost including opportunity cost
  • Urgency: Critical / High / Normal — how urgent is it?
  1. Call score_cr
  2. Receive:
  • A numeric CR Score and a preliminary verdict (Approve/Modify/Defer/Reject)
  • A narrative rationale from Claude that accounts for context
  • Automated checks (traceability to the need, conflicts with priorities)

Scoring scale:

  • ≥ 8.0 → ✅ Approve
  • 4.0–7.9 → 🟡 Modify
  • 1.0–3.9 → ⏳ Defer
  • < 1.0 → ❌ Reject
📌 The formula yields a preliminary verdict. Claude can adjust it
with an explicit reason. The BA makes the final decision on what gets submitted for approval.

For details on the axes: references/cr_assessment_guide.md → "Five impact analysis axes"


Step 4 — Record the decision (resolve_cr)

When: the authorized stakeholder has made a decision.

Algorithm:

  1. Obtain the decision from the authorized party (per governance 3.3 — score_cr

names the planned decision makers in its Step 4 block when 3.3 has been planned, and resolve_cr cross-checks decided_by against them and flags a mismatch in the CR Decision Record; it never blocks and never rewrites the name)

  1. Call resolve_cr with parameters:
  • decision: Approved / Approved_with_Modification / Deferred / Rejected
  • decided_by: who made the decision
  • rationale: justification (required — for audit purposes)
  1. The tool automatically:
  • On Approved: changes the status of affected requirements to under_change
  • Generates a CR Decision Record (Markdown) → save_artifact
  • Updates the status of the CR node in the 5.1 repository
  1. After Approved: update the content of requirements via update_requirement (5.2)
  2. After Approved: run start_prioritization_session (5.3) if priorities are affected
  3. Send the CR Decision Record via prepare_communication_package (4.4)
📌 Rejected and Deferred CRs are not deleted from the repository.
CR history is part of the project's audit trail.

MCP tools

ToolStepWhat it does
open_cr1Register the CR in the 5.1 repository
run_cr_impact2BFS traversal of the graph, build the list of affected items, create modifies relations
score_cr3Calculate the CR Score, get a recommendation + automated checks
resolve_cr4Record the decision, update statuses, generate the Decision Record

Mapping from 5.1 — relation types during impact analysis

Relation typeWhat it means for a CR
dependsThe dependent requirement may lose its meaning
verifiesTest cases need to be reviewed/rewritten
satisfiesThe code component needs to be reworked
derivesChild requirements may inherit the change
modifiesDirect CR → changed-requirement link (created in step 2)

Mapping from 5.2 — stability as a risk factor

Stability (from 5.2)Behavior in 5.4
Stable (version < 1.3)No additional flags
Volatile (version 1.3)🟡 Warning: unstable requirement + CR = double uncertainty
Volatile (version ≥ 1.4)🔴 High risk: recommend stabilizing the requirement before the CR

Common BA questions

"The CR is small — do I still need to go through all four steps?" Yes — all four steps. For a small CR, the pipeline moves through quickly. Impact often turns out larger than expected — the graph will show it.

"Who can be `decided_by`?" Determined in task 3.3 (Governance). Usually: the sponsor, the Product Owner, or the CCB. If unclear, check with the Project Manager.

"What if several CRs affect the same requirements?" Assess them together: pass related_cr_ids to open_cr. The combined impact can be non-linear — it's better to see the whole picture.

"After a CR is Approved, when should I update the requirements in 5.2?" Right after resolve_cr. The under_change status means "the requirement is in the process of being changed." Don't leave requirements in under_change for longer than one iteration.

"Do I need to re-prioritize after a CR?" If the CR changes Must/Won't items or adds new requirements — yes, run 5.3. resolve_cr will explicitly flag it if conflicts with priorities are detected.

Security

  • Security scanPassed, 100 of 100Pattern checks on the skill's files.
  • Model reviewNot runA model review is scheduled for the next rescan.

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

7-day trend score
0

Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/chaussky/ainalyst/requirements-assess-changes/badge)](https://verified-skill.com/skills/chaussky/ainalyst/requirements-assess-changes)

Something wrong with this skill or its page? Tell us.