agentleFS
envisioningenvisioning/skills/skills/merge/SKILL.md

merge

Skill2 starsChanged 51 days agoMITView on GitHub

Deduplicate a set of signal cards — find near-duplicates that describe the same underlying development, keep the most specific title, preserve the stronger evidence, and freeze the duplicates with a pointer instead of deleting them. Use when the user says "dedupe these", "merge duplicates", "these overlap", "clean up the scan", or after generating cards from several sources. Stage: structure.

54 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use addskill, apply-firmware 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.
  • ryanportfolioSkill

    addskill

    Create, import, update, or install repo or personal skills: runtime ownership, resources, discovery checks.

    Also in 6 projects that use this

  • ryanportfolioSkill

    apply-firmware

    Bring current folder (plain folder, existing repo, old or partial firmware copy) up to full Harness Firmware in place. Mirror external repo → adopt-repo.

    Also in 6 projects that use this

  • +10 more from ryanportfolio

The file

We split this skill into the rules it gives an agent

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

  • 0

    Common

    Most of the 48 edited versions we read keep these.

  • 0

    Distinctive

    This author's own choices. Worth a look.

  • 3

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  • could be automatic: a setting could enforce it
  1. Very specific:

    Read first: ../../core/signal-schema.md, ../../core/envisioning.md.

    line 11
  2. Very specific:

    Domain rule: If the cards carry a playbook, apply that playbook's *merging* line for what "same" means (same competitor + move.

    line 25
  3. Very specific:

    Never merge a verified card into a rejected one, or vice versa.

    line 54

“Common” is measured: kept by at least 6 in 10 of the 48 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

merge

Two cards that point at the same development are one signal with two titles. This skill finds them and folds them without losing the trail.

Read first: ../../core/signal-schema.md, ../../core/envisioning.md.

Input

A set of cards: a scan index, a folder, a list of paths or doors. Cards already carrying merged_into are skipped.

Procedure

  1. Pair. Compare cards on the What section, not the title. Two cards are candidates when they describe the same underlying development — the same programme, rule, launch, deal, capability, or measured shift. The same topic is not enough: "EU AI Act enforcement begins" and "EU AI Act fines first company" are two signals.
  2. Domain rule. If the cards carry a playbook, apply that playbook's merging line for what "same" means (same competitor + move; same docket; same incident; same one-hop opportunity).
  3. Choose the target. The card with the most specific, concrete title free of marketing, consulting, advocacy or alarmist framing. Ties → the better-grounded one.
  4. Fold. Into the target: the stronger evidence hook in the summary, the union of sources[] (dedupe by URL, keep the higher-confidence verdict), the union of actors[] and tech[]. Keep the target's category and metrics; if the merged-from card was scored higher on a metric, note it in Watch for rather than silently raising the score.
  5. Freeze. Each merged-from card keeps every field and gains merged_into: <target slug>. It is not deleted, not edited otherwise. Unmerging is removing that one line.
  6. Ask before folding when unsure. Present the pair with the sentence that makes them the same; a human decides. Never merge more than the user can review in one screen without pausing.
  7. Persist. Target and frozen cards per envisioning.md. message: merge · <n> into <target>.

Output

A table: target · merged-from · why they are the same. Then the count: <n> cards → <m> active, <k> frozen.

Refusals

  • Cards from different scopes or playbooks are not merged; the same fact can be a different signal to a different audience.
  • Never merge a verified card into a rejected one, or vice versa; that is a grounding disagreement to resolve with ground, not a duplicate.

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.