agentleFS
GustavoGutierrezGustavoGutierrez/engineering-skills/skills/traceability-manager/SKILL.md

traceability-manager

Skill0 starsChanged 5 months agoMITView on GitHub

Trigger: traceability, requirements traceability, forward traceability, backward traceability, change impact analysis, trace requirements to implementation, RTM, orphan detection, coverage model. Maintain bidirectional requirement-to-implementation traceability with typed links, coverage checks, impact analysis, and governance.

The file

We split this skill into the rules it gives an agent

12 rules. About 0 are common, 10 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈10

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this skill when the user needs to:

    line 26
  2. Distinctive:

    Do NOT use this skill for static spreadsheet management, project management tooling configuration, or business context documentation.

    line 35
  3. Distinctive:

    Use this controlled set of link types:

    line 50
  4. Distinctive:

    Every requirement must have at least one forward path to a validation artifact.

    line 64
  5. Very specific:

    Use glob and grep across the project to find relevant files with known extensions (*.md, *.spec.md, *.prd, *.prp, *.api.md).

    line 89
  6. Distinctive:

    When a requirement or artifact changes, trace its forward and backward links.

    line 108
  7. Distinctive:

    When the user requests a traceability plan (vs.

    line 199
  8. Distinctive:

    Before presenting the traceability model or plan, verify:

    line 244
  9. Distinctive:

    All artifacts found in scope are listed in the inventory.

    line 247
  10. Distinctive:

    Every relationship uses a defined typed link from the vocabulary.

    line 248
  11. Distinctive:

    No project-specific names, client names, or unnecessary vendor references were invented.

    line 252
  12. Very specific:

    All unknowns are marked as TBD rather than invented.

    line 254

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Traceability Manager

When to Use

Use this skill when the user needs to:

  • Establish or audit traceability between requirements and implementation artifacts.
  • Detect orphans: requirements without coverage, tasks without parents, APIs without source requirements.
  • Perform change impact analysis before modifying scoped requirements or artifacts.
  • Build or validate a Requirements Traceability Matrix (RTM).
  • Govern a traceability model: define typed links, coverage rules, and governance policies.
  • Generate a traceability plan or model in Markdown format.

Do NOT use this skill for static spreadsheet management, project management tooling configuration, or business context documentation. It produces executable traceability artifacts and is designed for both human review and agent-driven workflows.

Core Concepts

Bidirectional Traceability

Traceability flows in two directions:

  • Forward traceability: source requirement → design → code → tests → validation
  • Backward traceability: artifact → source requirement (confirms every implemented item is justified)

Bidirectional traceability enables both coverage analysis (all requirements are implemented) and impact analysis (a change to any artifact can be traced back to its origin).

Typed Link Vocabulary

Use this controlled set of link types:

Link TypeDirectionMeaning
originates-fromartifact → sourceThis artifact is derived from the named source artifact
refineslower-level → higher-levelThis requirement or task implements or clarifies the parent
implementstask → requirementThis task delivers the specified requirement
validated-byrequirement → testThis requirement is verified by the named test or validation
depends-onany → anyThis artifact requires the named artifact to be complete first
affectsany → anyA change to this artifact will impact the named artifact
governed-byany → policyThis artifact must comply with the named policy or standard

Coverage Model

Every requirement must have at least one forward path to a validation artifact. The coverage model distinguishes:

Coverage LevelRequired Links
uncoveredNo forward path to any validation artifact
partially-coveredForward path exists but not all acceptance criteria are validated
fully-coveredFull forward path from requirement to acceptance criteria to validation

Orphan Detection

Orphan types this skill detects:

Orphan TypeDefinition
orphan-requirementRequirement with no forward links to tasks or validation
orphan-taskTask with no backward link to a parent requirement or spec
orphan-apiAPI endpoint or contract not linked to a source requirement or spec
orphan-validationTest or validation artifact not linked to a traced requirement
orphan-policyArtifact with no governed-by link to a policy or standard

Operating Workflow

Phase 1: Discover Artifacts

  1. Identify the artifacts in scope: PRDs, PRPs, specs, stories, tasks, APIs, architectures, validation artifacts.
  2. Use glob and grep across the project to find relevant files with known extensions (*.md, *.spec.md, *.prd, *.prp, *.api.md).
  3. Build a preliminary inventory: artifact ID, type, file path, title or summary.

Phase 2: Map Relationships

  1. For each artifact, inspect content to identify explicit and implicit relationships.
  2. Apply the typed link vocabulary to record each relationship.
  3. Flag artifacts that lack explicit links as candidates for orphan detection.
  4. Identify governed-by relationships by checking for embedded policy references.

Phase 3: Coverage Analysis

  1. For each requirement or spec, check for a forward chain to tasks and validation.
  2. Classify each requirement into fully-covered, partially-covered, or uncovered.
  3. For each validation artifact, verify it has a backward link to a traced requirement.
  4. Record uncovered requirements and orphan artifacts.

Phase 4: Impact Analysis

  1. When a requirement or artifact changes, trace its forward and backward links.
  2. Identify all artifacts affected in both directions.
  3. For each affected artifact, assess the scope of the change.
  4. Present the impact set ranked by proximity from the changed artifact.

Phase 5: Produce Traceability Model

Generate the traceability model as a Markdown document using the required output structure.

Decision Rules

SituationAction
Requirement has no forward linkFlag as orphan-requirement; do not implement until parent is identified
Task has no backward linkFlag as orphan-task; ask for parent requirement or mark as infrastructure
API has no source requirementFlag as orphan-api; require a refines link to a requirement or spec before implementation
Validation has no backward linkFlag as orphan-validation; require a validated-by link to a requirement
Artifact has no policy linkFlag as orphan-policy; suggest a governed-by link to an applicable standard
Change to a requirementTrace forward and backward; present full impact set before proceeding
Multiple artifacts share a depends-onConsolidate the dependency chain; identify single source of truth
Conflicting links detectedSurface the conflict; do not resolve silently; present both paths for human decision

Traceability Model Output Structure

# Traceability Model

## 1. Overview
- Total artifacts:
- Requirements:
- Tasks:
- APIs:
- Validation artifacts:
- Policies:

## 2. Artifact Inventory
| ID | Type | Title | File | Links Out | Links In | Coverage |
| --- | --- | --- | --- | --- | --- | --- |

## 3. Relationship Graph
| From | Link Type | To | Rationale |
| --- | --- | --- | --- |

## 4. Typed Link Definitions
| Link Type | From | To | Cardinality | Coverage Rule |
| --- | --- | --- | --- | --- |

## 5. Coverage Report
### Fully Covered
| Requirement | Forward Path | Validation |
| --- | --- | --- |

### Partially Covered
| Requirement | Missing Link | Recommendation |
| --- | --- | --- |

### Uncovered (Orphans)
| Requirement | Gap | Recommended Action |
| --- | --- | --- |

## 6. Orphan Report
| Orphan Type | ID | Title | File | Recommended Action |
| --- | --- | --- | --- | --- |

## 7. Change Impact Analysis (when triggered)
### Changed Artifact
- ID:
- Type:
- File:

### Forward Impact (downstream)
| Artifact | Distance | Change Scope |
| --- | --- | --- |

### Backward Impact (upstream justification)
| Artifact | Distance | Rationale |
| --- | --- | --- |

## 8. Governance
- Link integrity rules:
- Minimum coverage standard:
- Policy compliance requirements:

## 9. Recommendations
- Immediate:
- Short-term:
- Long-term:

Traceability Plan Output Structure

When the user requests a traceability plan (vs. a snapshot model), use this structure:

# Traceability Plan

## 1. Objectives
- What the traceability model must enable

## 2. Scope
- Artifacts included
- Artifacts excluded
- Boundaries

## 3. Typed Link Architecture
- Defined link types
- Directionality rules
- Cardinality constraints

## 4. Coverage Expectations
- Minimum coverage level per artifact type
- Coverage gap threshold for blocking

## 5. Governance Rules
- Who approves link additions
- Orphan resolution SLAs
- Change impact review triggers

## 6. Workflow Integration
- When to update traceability links
- Who is responsible for each artifact type
- Review cadence

## 7. Reporting
- Coverage reports: frequency and audience
- Orphan reports: frequency and owner
- Impact analysis triggers and audience

## 8. Maintenance
- Link hygiene rules
- Orphan detection schedule
- Coverage recalculation triggers

Quality Checklist

Before presenting the traceability model or plan, verify:

  • The output is written in English.
  • All artifacts found in scope are listed in the inventory.
  • Every relationship uses a defined typed link from the vocabulary.
  • Uncovered requirements are listed separately with recommended actions.
  • Orphan artifacts are classified by orphan type with recommended actions.
  • Change impact analysis covers both forward and backward directions.
  • No project-specific names, client names, or unnecessary vendor references were invented.
  • The model specifies coverage expectations and governance rules.
  • All unknowns are marked as TBD rather than invented.

Files to Read

  • references/link-vocabulary.md — full typed link definitions with usage examples (load only when the skill body instructs it)

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.