# Sdrf:improve

> Use when the user wants to improve an existing SDRF file strictly according to SDRF templates and specification rules (no speculative additions).

- **Type:** Skill
- **Install:** `agentstack add skill-bigbio-sdrf-skills-sdrf-improve`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [bigbio](https://agentstack.voostack.com/s/bigbio)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [bigbio](https://github.com/bigbio)
- **Source:** https://github.com/bigbio/sdrf-skills/tree/main/skills/sdrf-improve
- **Website:** https://sdrf.quantms.org

## Install

```sh
agentstack add skill-bigbio-sdrf-skills-sdrf-improve
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# SDRF Specification-Driven Improvement Workflow

You are improving an SDRF file using ONLY specification/template rules.

Do not suggest additions based on:
- Similar datasets in PRIDE
- Literature expectations not encoded in templates
- "Could be more detailed" heuristics
- Personal curator preference

If a change is not justified by template metadata (`required`/`recommended`) or
TERMS.tsv rules, do not recommend it.

## Authoritative Sources (must read)

1. `spec/sdrf-proteomics/TERMS.tsv`
2. `spec/sdrf-proteomics/sdrf-templates/templates.yaml`
3. `spec/sdrf-proteomics/sdrf-templates/{name}/{version}/{name}.yaml` for each template in use

For affinity-proteomics, align with the official template/spec rules:
- Technology template: `affinity-proteomics`
- Optional experiment child template: `olink` OR `somascan` (mutually exclusive)

## Step 1: Determine Active Templates

1. Parse `comment[sdrf template]` columns (NT/VV format).
2. If missing/incomplete, detect from SDRF content and technology type.
3. Confirm template set with the user before proposing edits.

Do not "upgrade" template versions automatically. If a newer version exists, report it as an optional migration task.

## Step 2: Build Rule Matrix

Construct an explicit checklist from template YAML files:
- Required columns
- Recommended columns
- Optional columns
- Column validators (values/patterns/ontology)
- Allowed reserved words (`not available`, `not applicable`, `pooled`) via TERMS.tsv flags

For affinity-proteomics specifically, ensure checks include:
- `comment[platform]` (required)
- `comment[panel name]` (recommended)
- `comment[quantification unit]` (optional; values include NPX/RFU in spec)
- `comment[normalization method]` (optional)
- `comment[fraction identifier]` (optional)

And for child templates:
- Olink: check Olink-specific required/recommended columns from `olink.yaml`
- SomaScan: check SomaScan-specific required/recommended columns from `somascan.yaml`

## Step 3: Identify Spec-Backed Improvements

Classify findings into these categories only:

### A. Required Fixes (must change)
- Missing required columns from active templates
- Invalid column names not matching SDRF naming patterns
- Values violating template/TERMS validators
- Invalid reserved words per TERMS flags

### B. Recommended Fixes (should change)
- Missing columns marked `recommended` in active templates
- Values that fail recommended validators (warnings)

### C. Optional Enhancements (may change)
- Missing optional columns from active templates
- Only include if explicitly present in template metadata

Do not include free-form "quality" recommendations outside A/B/C.

## Step 4: Generate Deterministic Report

Report every finding with source traceability:
- Column or value issue
- Severity (`required` / `recommended` / `optional`)
- Exact source rule:
  - template file + column definition, or
  - TERMS.tsv field (usage/values/allow_not_available/allow_not_applicable/allow_pooled)
- Proposed correction

Example format:
```text
Finding: Missing column `comment[panel name]`
Severity: recommended
Source: spec/sdrf-proteomics/sdrf-templates/affinity-proteomics/1.0.0/affinity-proteomics.yaml
Action: Add `comment[panel name]` with panel identifier values.
```

## Step 5: Apply Changes Only with User Approval

Before modifying file contents:
1. Show the exact changes (old -> new)
2. Group by severity (required first)
3. Ask user approval for recommended/optional changes

Required fixes can be applied directly if the user asked to "fix all required issues."

## Output Constraints

- No speculative metadata additions
- No PRIDE peer comparison suggestions
- No literature-derived additions unless already required/recommended by templates
- No ontology "more specific child" suggestions unless validator explicitly requires a value constraint

The goal is strict conformance improvement, not curation enrichment.

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [bigbio](https://github.com/bigbio)
- **Source:** [bigbio/sdrf-skills](https://github.com/bigbio/sdrf-skills)
- **License:** MIT
- **Homepage:** https://sdrf.quantms.org

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-bigbio-sdrf-skills-sdrf-improve
- Seller: https://agentstack.voostack.com/s/bigbio
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.