Claude Skill MartBrowse skillsQuick linesLearn by videoTerminal guideWhat is a Skill?
 
← Back to list

FDE Engagement Record (@fde)

A consulting-operations skill that routes client work through six stages and keeps every agreement, decision, and piece of evidence in a local `.fde/` engagement record.

AutomationAdvanced★ 534⑂ 118AI score 9/10Last updated: Sep 14, 2026

What it does

  • Routes client work across six stages (Land → Discover → Plan → Ship → Outcome → Close). You never pick a skill or phase — the agent listens for cues ("taking over a project", "let's POC this", "the sponsor went quiet") and loads exactly one reference doc.
  • Turns pasted meeting notes into a reviewed update: fde debrief --smart → agent reconciliation → one plain-English review → save only after you confirm.
  • Answers "when did we agree to that?" from the record instead of memory (fde receipts). Claims without a supplied source stay CLAIM; sourced ones become ON RECORD.
  • Enforces a delivery ground loop: name the plan, characterise their existing code, verify in the agreed environment, authorize release, then log outcome — plus a rehearsed rollback path and explicit implemented / verified / deployed / accepted status.
  • Includes data-boundary rules: local-only CLI, identifier masking, <private> blocks that model-facing file tools must not open.

Who it's for

  • Forward-deployed engineers, solutions engineers, and technical consultants working inside a client's environment.
  • Anyone juggling multiple clients who keeps losing context between engagements.
  • Teams where green CI isn't "done" — you need a named signer and replayable acceptance evidence.

Example uses

  1. Post-meeting debrief: "just came out of the Acme call" + pasted notes → the agent drafts the update, surfaces only consequential changes, asks "Save this update?", then writes .fde/.
  2. Scope dispute: "what did we agree about the payment fee?" → runs fde receipts and answers with dates and sources.
  3. Friday readout: "I need to send the sponsor something" → fde status plus the readout reference, separating promised vs measured vs accepted.

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/suboss87/FDEOps/HEAD/skills/fde/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on Claude Skill Mart.
Copy the skills/fde folder from the GitHub repo suboss87/FDEOps into my ~/.claude/skills/fde/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/suboss87/FDEOps.git && mkdir -p ~/.claude/skills && cp -r FDEOps/skills/fde ~/.claude/skills/fde

⚠ This is a third-party skill. Check the source repository before installing.

  1. Open a terminal.
  2. Clone the repo: git clone https://github.com/suboss87/FDEOps.git
  3. Copy the skill into your Claude Code skills folder: mkdir -p ~/.claude/skills && cp -r FDEOps/skills/fde ~/.claude/skills/fde
  4. This skill drives the fde CLI. With Node.js 18+ installed, the agent can fall back to npx --yes fdeops ...; verify once with npx --yes fdeops --version.
  5. Restart Claude Code, open the working folder for that client, and say @fde or just describe the situation in plain English.
  6. On first use you'll be asked one question — what to call the client — after which the agent runs fde resume --init itself to create the .fde/ record.
  7. Optional: run fde setup locally to set working style and masking terms. Enter sensitive names in the CLI or via --terms-file, never by pasting them into chat.
View source on GitHub ↗License: MIT