AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

Osx Modify Artifacts

skill-amauryconstant-openspec-extended-osx-modify-artifacts · by amauryconstant

Modify an existing artifact in an OpenSpec change. Use during pre-implementation review iteration (after review-artifacts feedback) or during implementation when requirements change. Supports natural language descriptions and targeted edits with dependency tracking.

No reviews yet
0 installs
28 views
0.0% view→install

Install

$ agentstack add skill-amauryconstant-openspec-extended-osx-modify-artifacts

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-amauryconstant-openspec-extended-osx-modify-artifacts)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
3mo ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Osx Modify Artifacts? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Modify an existing artifact in an OpenSpec change with dependency-aware updates.

IMPORTANT: context and rules from openspec instructions are constraints for YOU, not content for the artifact file. Do NOT copy `, , or ` blocks into artifacts. These guide what you write but should never appear in output.


Input

Optionally specify a change name and artifact ID. If omitted, the skill will infer from context or prompt for selection.

Arguments: [change-name] [artifact-id]

Examples:

  • /osx-modify add-auth - Modify change "add-auth" (will prompt for artifact)
  • /osx-modify add-auth proposal - Modify proposal.md in "add-auth"
  • "Fix the missing rationale in design decision 2" - Infer from context

Steps

  1. Select the change

If a name is provided, use it. Otherwise:

  • Infer from conversation context if the user mentioned a change
  • Auto-select if only one active change exists
  • If ambiguous: run openspec list --json and use the AskUserQuestion tool to let the user select

When showing changes, include: name, schema, status, last modified. Mark the most recently modified as "(Recommended)".

Always announce: "Using change: " and how to override (e.g., /osx-modify ).

  1. Check change status

``bash openspec status --change "" --json ``

Parse the JSON to understand:

  • schemaName: The workflow being used (e.g., "spec-driven")
  • artifacts: Array of artifacts with status (done/ready/blocked)
  • isComplete: Whether all artifacts are done
  1. Select artifact to modify

If an artifact ID is specified, use it. Otherwise:

  • If only one artifact has status "ready": auto-select it
  • If user described content (e.g., "the requirements", "the design"): match by name
  • If multiple artifacts ready and no direction: use the AskUserQuestion tool to prompt

When prompting, present artifacts in schema order, showing:

  • Artifact ID
  • Status (done/ready/blocked)
  • Dependencies count
  • Unlocks count
  1. Get modification context

``bash openspec instructions --change "" --json ``

Parse the JSON to extract:

  • rules: Validation rules for this artifact (constraints for you)
  • context: Project background (constraints for you)
  • template: Expected structure
  • dependencies: Artifacts this artifact depends on
  • unlocks: Artifacts that depend on this one
  • outputPath: Where the artifact file is located
  • instruction: Schema-specific guidance

Read the current artifact file from outputPath.

  1. Detect workflow scenario

Check conversation context for review feedback:

  • If user references review-artifacts output or specific issue numbers → Review Iteration Mode
  • If user describes new requirements or changes discovered during coding → Amendment Mode

Review Iteration Mode:

  • Focus on specific issues identified in review
  • Validate fixes against the review criteria
  • After modification, suggest re-running review-artifacts

Amendment Mode:

  • Focus on capturing requirement changes
  • Consider impact on existing implementation
  • After modification, suggest running /osx-apply
  1. Display validation constraints

Show the user:

  • rules array (as constraints, not content)
  • dependencies list (what this artifact relies on)
  • unlocks list (what artifacts depend on this)
  1. Determine modification mode

Mode A: Describe Changes - Use when user provides natural language:

  • "Add a requirement for..."
  • "Update the design to..."
  • "Remove this section..."

Mode B: Interactive Edit - Use when user references specific content:

  • "Change line 42 to..."
  • "Replace the second paragraph..."
  • "Update the authentication section..."

Auto-select based on input type. No explicit prompt needed.

  1. Apply modifications

Mode A (Describe Changes):

  • Parse user's description
  • Analyze current artifact content
  • Identify which sections need changes
  • Apply changes autonomously if clear
  • If ambiguous: pause and ask for clarification using AskUserQuestion tool

Mode B (Interactive Edit):

  • Parse entire artifact file
  • Identify relevant sections based on user's edit intent
  • Show only those sections (not the entire file)
  • User provides specific edit instructions
  • Apply targeted changes using Edit tool
  • If more sections need changes, repeat
  1. Validate modifications

Check proposed changes against:

a) The rules array from step 4 (structural/format requirements)

b) If in Review Iteration Mode: The specific issues from review-artifacts feedback

c) If in Amendment Mode: Backward compatibility with implementation

Handle validation results:

  • Clear/fixable violations → Fix automatically and continue
  • Ambiguous violations → Explain issue and ask user
  • If user's intent is clear despite violation → Proceed with warning

Note: Use rules from instructions. Do NOT run openspec validate.

  1. Write the updated artifact file

Use Edit tool for targeted changes, Write tool for complete rewrites. Verify the file was written successfully.

  1. Handle dependent artifacts

From the instructions output, check the unlocks array (reverse dependencies).

For each artifact in unlocks:

  • Run openspec instructions --change "" --json
  • Read the dependent artifact file
  • Analyze if the modification affects this dependent artifact
  • Track affected artifacts

Decision logic (prefer reasonable decisions):

  • Single dependent affected: Auto-update and explain (no prompt)
  • Multiple dependents affected: Show the list and prompt for confirmation
  • User mentioned "cascade": Auto-update regardless of count
  1. Show success summary

Display:

  • Which artifact was modified
  • Changes applied (summary)
  • Dependent artifacts updated
  • Context-aware next steps based on scenario

Output

On Success:

## Modification Complete

**Change:** 
**Artifact:** 
**Mode:** [Review Iteration / Amendment]

### Changes Applied
- [Section]: [Action] - [Summary]

### Dependent Artifacts Updated
- [x] : [Summary]

### Next Steps
**Pre-implementation review:**
- Continue with: [next flagged artifact if any]
- Verify fixes: `/osx-review `
- Ready to implement: `/osx-apply ` after review passes

**Implementation phase**:
- Sync changes: `/osx-apply `
- Check tests: [affected test files if known]

On Pause (Ambiguous Input):

## Modification Paused

**Issue:** 

**Options:**
1. 
2. 
3. Describe differently

What would you like to do?

Guardrails

  • Always read current artifact before modifying
  • Check dependents before finalizing changes
  • Use rules from instructions for validation (not CLI validate command)
  • Use Edit for targeted changes, Write for complete rewrites
  • Prefer reasonable decisions to keep momentum (0-1 dependents → auto-update)
  • Pause and ask for clarification if unable to act autonomously
  • Follow schema order for artifact selection
  • Never copy context, rules, or project_context blocks into artifact files

Source & license

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

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

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.