- Home
- Skills
- Code Quality & Review
- scope-map
More screenshots
Works with the AI Agents you already use
Scope Map
Index codebases into low-token maps to trace bugs, define strict task scopes, and audit git diffs against drift.
$39
scope-map
Example session with this skill installed
Generate a map for the current directory, then query the PaymentGateway class to see its dependencies before I refactor the process_refund method.
- Read your context and instructions
- Compiled the scope-map
I have indexed the codebase and retrieved the dependency mapping for PaymentGateway.
Code Map Query: PaymentGateway
| Component | Type | File Path | Line Range | Dependencies |
|---|---|---|---|---|
PaymentGateway | Class | src/services/billing.py | 42-185 | Database, Logger, StripeClient |
process_refund | Method | src/services/billing.py | 112-140 | StripeClient::void_transaction |
Dependency Analysis
- Used By:
OrderController::cancel,AdminPortal::process_return - Calls:
StripeClient::void_transaction(Line 125),Database::update_record(Line 138) - Evidence:
script(AST parsed)
The process_refund method is tightly coupled with StripeClient. Before refactoring, ensure that OrderController is updated to handle the new return signature. I have prepared the scope for a new trace.
Next steps
- Run
python3 scripts/code_map.py trace map.json trace-refactor-001.jsonto define the refactor scope. - Open
map.htmlto visualize the call chain betweenOrderControllerandPaymentGateway. - Use the Copy agent instruction button in the viewer to delegate the specific method update.
Connects securely to your tools. The creator never sees your data.
About this skill
Stop wasting tokens and context window on sprawling codebases. When debugging or refactoring, most agents waste cycles reading files they do not need. This skill forces a "locate first, read second" workflow, using a lightweight map to identify exact line ranges before opening a single file. It provides a source of truth for humans via an HTML viewer and for agents via low-token queries.
What it does
- Map generation creates a
map.jsonindexing modules, classes, and methods across major languages using AST or regex. - Dependency tracking identifies "uses/used-by" and "calls/called-by" relationships without loading full source files.
- Trace recording documents bug symptoms and step-by-step call chains into a structured JSON format.
- Scope auditing compares
git diffagainst declared task scopes to prevent accidental edits in unrelated files. - Agent delegation generates explicit instructions for subagents, restricting them to specific line ranges.
How it works
- Generate the project map using
code_map.pyto index the codebase structure and dependencies. - Search and Query the map to follow call chains from a symptom to a suspect component in ~100 tokens.
- Trace the path in a trace file, defining exact line-level scopes for the fix or feature.
- Audit the final changes against the trace to ensure no unauthorized files or methods were modified.
Frameworks & tools
Built with Python 3 using the Standard Library. Supports Python (AST), PHP, JS/TS, Java, C#, Kotlin, Go, Ruby, and GDScript. Integrates with Git for scope auditing.
Why this beats prompting it yourself
Generic prompts often lead to "hallucinated" file paths or massive context dumps that swallow your budget. This skill enforces a strict verification layer where every task is scoped to specific lines, and unauthorized changes are caught by a hard audit check before they are committed.
Use cases
- Debugging complex regressions by tracing call stacks through unfamiliar modules.
- Mapping dependencies before a major refactor to see the "blast radius" of changes.
- Delegating specific tasks to subagents with strict line-level boundaries.
- Auditing PRs or agent outputs to ensure zero scope creep.
Known limitations
Regex extraction for non-Python languages may omit dynamic or overloaded relationships. Requires a local Python 3 environment to run the mapping script.
Regex-based extraction for JS/TS, PHP, and others may miss dynamic links. Python environment required. Exits with error if git diff drifts from declared scope.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 3 days ago
Requires