agentleFS
karim-bhalwanikarim-bhalwani/agent-skills-collection/skills/navigator/SKILL.md

navigator

Skill0 starsChanged 10 months agoMITView on GitHub

Specialized in codebase exploration, finding file paths, tracing execution logic, and mapping dependencies. Use when exploring unfamiliar codebases, finding where code is located, tracing function calls, understanding module structure, or mapping system dependencies.

3 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

8 rules. About 0 are common, 6 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    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. Very specific:

    Discovery: Use find_by_name and list_dir to map directory structures.

    line 23
  2. Very specific:

    Symbol Search: Use grep_search to find classes, functions, or specific strings across the project.

    line 24
  3. Distinctive:

    Every navigation task should summarize:

    line 42
  4. Distinctive:

    When entering a new or unfamiliar codebase.

    line 51
  5. Distinctive:

    Before implementation to verify "where to cut" into the code.

    line 52
  6. Distinctive:

    When trying to find the source of a specific error or behavior.

    line 53
  7. Distinctive:

    DO NOT hallucinate paths: Only report what tools find.

    line 66
  8. Distinctive:

    Missing File References: Not tracing all imports means some related files are missed.

    line 71

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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

Navigator Skill - Filesystem & Logic Discovery

Overview

The Navigator skill is the "GPS and Surveyor" of the codebase. It helps understand "where things are" and "how they work" before any changes are made.

Core Capabilities

1. Filesystem Scouting (Locator)

  • Discovery: Use find_by_name and list_dir to map directory structures.
  • Symbol Search: Use grep_search to find classes, functions, or specific strings across the project.
  • Context Extraction: Identify relevant files for a specific feature or bug without reading entire contents.

2. Logic Analysis (Analyzer)

  • Execution Tracing: Map data flows from entry points to persistence.
  • Dependency Analysis: Identify what a module relies on and what relies on it.
  • Integration Seams: Pinpoint the exact location where a new feature should be injected or a bug fix should be applied.

Workflow

  1. Locate: Find all files related to the user's request using naming conventions and content searches.
  2. Read & Trace: Ingest the relevant files and follow imports/calls to understand the chain of command.
  3. Explain: Summarize the logic in plain English for implementation planning.
  4. Handoff: Provide a clear list of files and specific line numbers for the next phase.

Output Format: Discovery Report

Every navigation task should summarize:

  • Relevant Files: List of absolute paths with a 1-sentence description of their role.
  • Logic Summary: Explanation of the current execution flow.
  • Dependency Map: Visual or text-based map of module interactions.
  • Recommended Seam: Where modifications should specifically take place.

When to Use

  • When entering a new or unfamiliar codebase.
  • Before implementation to verify "where to cut" into the code.
  • When trying to find the source of a specific error or behavior.

Outputs & Deliverables

  • Primary Output: Discovery report summarizing relevant files, logic summary, and recommended seam (plain-text or markdown)
  • Secondary Output: File list with line references and a short dependency map (Mermaid or ASCII)
  • Success Criteria: Report identifies the correct set of files and a clear recommended modification seam
  • Quality Gate: implementer or architect confirmation before changes are made

Constraints

  • NO implementation or refactoring.
  • NO writing to files.
  • DO NOT hallucinate paths. Only report what tools find.

Common Pitfalls

  • Incomplete Symbol Search: Searching for "user" finds hundreds of matches. Use more specific patterns and context to narrow results.
  • Missing File References: Not tracing all imports means some related files are missed. Follow the import chain completely.
  • Misreading Execution Flow: Assuming linear execution without understanding async/callbacks/events leads to wrong seams. Trace carefully.
  • Ignoring Circular Dependencies: Finding mutual imports indicates a design problem. Flag these as risks, not just connections.
  • Wrong Modification Point: Recommending changes at the wrong layer. Understand the full flow before suggesting "cut here".
  • Hallucinated Files: Inventing file paths that don't exist. Only report paths confirmed by directory listings and searches.

Integration Points

PhaseInput FromOutput ToContext
ExplorationUser question about codebaseDiscovery reportFind relevant files and logic flow
ValidationReport findingsimplementerConfirm modification seam is correct
ArchitectureSystem understandingarchitectProvide dependency map for design decisions
ImplementationReady-to-code stateimplementerList exact files and line numbers for changes

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.