agentleFS
priyansh1210priyansh1210/shiro/.claude/skills/setup/SKILL.md

setup

Skill1 starsChanged 5 days agoMITView on GitHub

First-time setup. Creates the private career memory, imports everything from the user's current resume, and collects the standard answers job application forms ask for.

192 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use add-voice-transcription, add-discord and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • juancruzmunozalbeloSkill

    add-voice-transcription

    Add voice message transcription to NanoClaw using OpenAI's Whisper API. Automatically transcribes WhatsApp voice notes so the agent can read and respond to them.

    Also in 5 projects that use this

  • MicrosoftSkill

    add-discord

    Add Discord bot channel integration via Chat SDK.

    Also in 11 projects that use this

  • MicrosoftSkill

    add-slack

    Add Slack channel integration via Chat SDK.

    Also in 11 projects that use this

  • juancruzmunozalbeloSkill

    add-telegram-swarm

    Add Agent Swarm (Teams) support to Telegram. Each subagent gets its own bot identity in the group. Requires Telegram channel to be set up first (use /add-telegram). Triggers on "agent swarm", "agent teams telegram", "telegram swarm", "bot pool".

    Also in 7 projects that use this

  • +8 more from microsoft

The file

We split this skill into the rules it gives an agent

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

  • 0

    Common

    Most of the 169 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Never overwrite an existing memory file: If memory already exists, tell the user and continue in update mode: only fill in what's missing.

    line 15
  2. Distinctive:

    Ask in short batches. Use multiple-choice questions where the answers are fixed.

    line 27

“Common” is measured: kept by at least 6 in 10 of the 169 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

/setup: build the career memory

Arguments: $ARGUMENTS

1. Create memory/

For each file in templates/memory/, create memory/<same name> if it doesn't exist yet. Also create the folders memory/resume/ and memory/documents/.

Never overwrite an existing memory file. If memory already exists, tell the user and continue in update mode: only fill in what's missing.

2. Import the resumes

People often keep one resume per kind of role (for example software engineer and data analyst). Import all of them.

  1. Take the path(s) from the arguments, or ask for them. Ask whether there are more resumes for other kinds of roles.
  2. Copy each file into memory/resume/ with a shell copy, naming it after the role it targets (e.g. software-engineer.pdf). If a file is already in memory/resume/ (for example dropped in the Shiro app), don't copy it; rename it in place with node tools/file-doc.mjs memory/resume/<file> resumes/<role-name> (one plain command, no cd). Add one line per resume under ## Resumes in memory/profile.md: - <Role>: memory/resume/<file>. Keep <Role> to a few words (e.g. AI Engineer Intern, or AI Engineer Intern (Acme) for a version made for one company), because the app shows it in its resume pickers; don't write notes about what's in the resume there, the picker reads the resume itself. Keep file names short too (role, plus the company or v2 when needed). Ask which role a resume targets if it isn't obvious.
  3. Read each one and move every fact into the right memory file, using the table in .claude/skills/remember/SKILL.md: contact details and links go to profile.md, each role to experience.md, and so on. When resumes describe the same job differently, keep every distinct fact under one entry. Keep the user's wording and numbers exactly. Don't improve anything at this stage.
  4. When resumes disagree (dates, titles, grades), ask the user which is right, and ask which resume is the most accurate overall. Record it as Reference resume under ## About this memory in profile.md, and set Last updated to today. From then on memory is the source of truth (see rule 5 in CLAUDE.md).

3. Fill the gaps

In the Shiro app you can't show multiple-choice pickers: write each batch as a short numbered list in your final message (offer the options inline) and stop; the user's reply continues the setup. Skip any question the user already answered.

Ask in short batches. Use multiple-choice questions where the answers are fixed. Everything can be skipped. Never guess.

  1. Contact and links that aren't on the resume: phone with country code, LinkedIn, GitHub, portfolio.
  2. Eligibility: countries where they can work, sponsorship needs, willingness to relocate, remote / hybrid / onsite preference.
  3. Availability and pay: current status, graduation date (if a student), earliest start date or notice period, expected compensation (or "prefer to discuss").
  4. Self-identification questions (gender, ethnicity, veteran, disability). Explain that these are voluntary on forms and "Decline to self-identify" is a fine answer for all of them.
  5. Form defaults: how they usually heard about jobs, languages, over 18, OK with a background check.
  6. What's not on the resume: other internships, courses, certificates, projects, hackathons, awards, leadership. Save each one following the /remember rules.
  7. Documents and links forms ask for: certificate PDFs or verification links, transcripts, degree certificate, recommendation letters, portfolio, writing or work samples. Ask for file paths and links, and save them following the "Files and links" rules in .claude/skills/remember/SKILL.md. Mention that they can also drop files into the Memory tab of the Shiro app (npm run ui). Don't accept identity or bank documents.

4. Finish

Summarize what's in memory in a few lines: how many entries per file, and anything still blank that forms commonly ask for. Then tell them:

  • /review <JD file or URL> runs the panel on a job.
  • /apply <application URL> fills an application (needs Chrome, see the README).
  • /remember <anything new> keeps memory current. Claude also saves new facts on its own when they mention them.

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.