agentleFS
envisioningenvisioning/skills/skills/frames/SKILL.md

frames

Skill2 starsChanged 51 days agoMITView on GitHub

Synthesize a set of signal cards into frames — named narrative groupings such as drivers of change, tensions, or emerging themes — each with a title, summary, body and the signals it binds, many-to-many. Use when the user asks "what are the themes", "drivers of change", "cluster these signals", "what's the story here", "frames", or needs the layer between raw signals and a report. Stage: structure.

2 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

3 rules. About 0 are common, 1 is distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    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:

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

    line 12
  2. Distinctive:

    Append, don't replace: If a frame set already exists for this scope, new frames are added alongside.

    line 46
  3. Very specific:

    No forecasts inside frames.

    line 76

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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

frames

A frame is a narrative that several signals belong to. Frames are how a scan becomes readable without losing the cards underneath: every claim in a frame points back to the signals that support it.

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

Input

Active signal cards (a scan index, folder, or list; frozen merged_into cards are skipped), and a frame kind. Default kinds:

kinda frame is…good count
driversa force pushing the scope in one direction4–7
tensionstwo forces pulling against each other3–5
themesa pattern several signals share5–8
betsa claim the corpus supports that a strategist could act on3–5

The user can name another kind; write its definition into the output.

Procedure

  1. Read every card. Frames come from the What and So what sections, not the titles.
  2. Propose frames. Each needs at least two signals; a one-signal frame is that signal restated. A signal may sit in several frames — that is the point of many-to-many, and a signal in three frames is usually the most important one in the corpus.
  3. Write each frame:
    • title — 3–6 words, a claim not a topic ("Regulators move before standards bodies", not "Regulation").
    • summary — two sentences, present tense.
    • content — 150–250 words: the mechanism, the actors, the evidence, with each signal cited inline by slug the first time it is used.
    • signals[] — the bound cards.
  4. Check coverage. List signals bound to no frame. Either they are noise (say so) or a frame is missing (add it). Don't force them in.
  5. Append, don't replace. If a frame set already exists for this scope, new frames are added alongside; existing ones are edited only when the user asks. Provenance is the version history.
  6. Persist to /foresight/frames/<scope>-<kind> (category: research), or ./foresight/frames/. message: frames · <kind> · <n> frames.

Output

---
kind: frames
frame_kind: <drivers|tensions|themes|bets|custom>
title: <scope> · <frame kind>
scope:
sources: [<scan index or cards>]
signals_total: <n>
signals_unbound: [<slugs>]
---
## <frame title>
<summary>

<content>

**Signals:** slug · slug · slug

Refusals

  • Fewer than 6 active cards → say a frame set won't be meaningful; offer to run a scan or signal first.
  • No forecasts inside frames. A frame describes what the signals show now; scenarios is where futures live.

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.