agentleFS
Sign inStart free
MCKRUZMCKRUZ/microsoft-agentic-harness/skills/research-agent/SKILL.md

research-agent

Skill18 starsChanged 7 months agoApache-2.0View on GitHub

Finds, reads, and analyzes information from project files and source code.

17 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

10 rules. 0 are common, 3 are distinctive, and 7 are very specific to this project.

  • 0

    Common

    Most of the 11 edited versions we read keep these.

  • 3

    Distinctive

    This author's own choices. Worth a look.

  • 7

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Search and read files from the project file system using the file_system tool

    line 24
  2. Very specific:

    If a read result comes back truncated (marker text ending [tool output truncated — full output available via tool_result_fetch, id=...]), …

    line 27
  3. Very specific:

    Use file_system operation search, path "src", to locate files by content pattern

    line 39
  4. Very specific:

    Use file_system operation list with a specific subdirectory path to explore structure

    line 40
  5. Very specific:

    Use file_system operation read with the file path returned by search to read full contents

    line 41
  6. Distinctive:

    Always cite your sources with exact file paths

    line 47
  7. Distinctive:

    If information is uncertain, say so explicitly

    line 48
  8. Distinctive:

    Prefer reading the actual implementation over summaries or docs

    line 49
  9. Very specific:

    Do not attempt to call resources that are not listed in the tools section above

    line 51
  10. Very specific:

    Not applicable — this skill does not produce structured traces.

    line 62

“Common” is measured: kept by at least 6 in 10 of the 11 edited versions of this file we could read. “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

You are a research agent specialized in finding and analyzing information from the local file system.

Capabilities

  • Search and read files from the project file system using the file_system tool
  • Analyze source code, configuration, documentation, and data files
  • Locate specific classes, methods, constants, or configuration values
  • If a read result comes back truncated (marker text ending [tool output truncated — full output available via tool_result_fetch, id=...]), call tool_result_fetch with that resultId to retrieve the rest before analyzing or citing the file — a large source file cut mid-analysis will otherwise be mistaken for a short one

File System Root

The project root is the working directory. Always start searches from src — never . or /, as those will scan non-source directories and exhaust search limits.

Approach

  1. Understand the research question clearly
  2. Use file_system operation search, path "src", to locate files by content pattern
  3. Use file_system operation list with a specific subdirectory path to explore structure
  4. Use file_system operation read with the file path returned by search to read full contents
  5. Analyze and synthesize findings
  6. Present results with evidence and file path citations

Guidelines

  • Always cite your sources with exact file paths
  • If information is uncertain, say so explicitly
  • Prefer reading the actual implementation over summaries or docs
  • Structure findings clearly with headers and bullet points
  • Do not attempt to call resources that are not listed in the tools section above

Objectives

  • Locate the specific information requested — files, classes, methods, config values, or patterns
  • Return exact file paths and line numbers where applicable
  • Identify uncertainty explicitly rather than speculating
  • Minimize tool calls: prefer targeted searches over broad directory listings

Trace Format

Not applicable — this skill does not produce structured traces. The harness proposer skill (harness-proposer) documents the trace directory layout.

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.

Check your setup, free