New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Conversation Compaction

by Prismer-AI in Prismer-AI/PrismerCloud

Produce the SESSION PROJECTION of memory for an aging conversation slice.

MITAuto-check passed

Install Conversation Compaction

skills CLI
$ npx skills add Prismer-AI/PrismerCloud --skill conversation-compaction -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install Prismer-AI/PrismerCloud conversation-compaction --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/Prismer-AI/PrismerCloud.git skills-src && mkdir -p .claude/skills && cp -r skills-src/sdk/cloud/catalog/skills/conversation-compaction .claude/skills/conversation-compaction && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
conversation-compaction
GitHub stars
1.6k
Token cost
~1.5k tokens
SKILL.md length
619 words
Files
1
Skills in repo
88
Repo updated
First seen
Licence
MIT

At a glance

Produce the SESSION PROJECTION of memory for an aging conversation slice.

  • SKILL.md covers The model — sync by POINTER,…, Input you receive, Pointer-first rule (recall-fed) and Flow, plus 4 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Conversation Compaction is an agent skill from Prismer-AI/PrismerCloud. Produce the SESSION PROJECTION of memory for an aging conversation slice. Runs as a single background model call (mirroring the memory extract hook) with NO live memory tools — durable capture to memory is extract's job. You MATCH the slice's durable facts to the memory pages you're handed (recall-found <existing-memory-pages) and emit POINTERS to them; your unique output is the thin ephemeral residue (open threads, abandoned directions) plus those pointers. A digest is memory's projection onto the session…

Its SKILL.md is about 1.5k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

The licence is MIT.

Example prompts

  • “s job. You MATCH the slice”
  • “/conversation-compaction”

What it can do on your machine

Read from SKILL.md and the folder at commit e5d9444. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md (its code samples are json).

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Conversation Compaction loads about 1.5k tokens when it runs. Until then it costs about 172 tokens; SKILL.md has 619 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~172
When it runs · the whole SKILL.md, loaded when a task matches
~1.5k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from Prismer-AI/PrismerCloud at commit e5d9444, republished under its MIT licence (© Prismer-AI). 619 words, ~1,527 tokens.

Download SKILL.mdSave it as .claude/skills/conversation-compaction/SKILL.md (or your agent's skills folder).
name
conversation-compaction
description
Produce the SESSION PROJECTION of memory for an aging conversation slice. Runs as a single background model call (mirroring the memory `extract` hook) with NO live memory tools — durable capture to memory is `extract`'s job. You MATCH the slice's durable facts to the memory pages you're handed (recall-found `<existing-memory-pages>`) and emit POINTERS to them; your unique output is the thin ephemeral residue (open threads, abandoned directions) plus those pointers. A digest is memory's projection onto the session, synced by POINTER, never by re-summarizing what memory already holds. Output is a compressed segment the dispatcher splices into future context.
scope
common
metadata.internal
true
metadata.excludeFromDefaultInstall
true

Conversation Compaction — memory's session projection

You are producing the session projection of a slice of an aging conversation (messages that scrolled past the recent verbatim window). The projection replaces the raw slice in future agent context.

The model — sync by POINTER, not by re-summary

Durable knowledge lives in memory (the memory skill owns it). A session projection does not re-summarize durable facts — it points at the memory pages that hold them, NOT by you re-deriving what memory already knows.

How this runs (important): you execute as a SINGLE model call inside the agent's daemon (mirroring the memory extract post-turn hook) — you have no live memory tools to browse/write. Durable capture to memory is the extract hook's job (it runs alongside you every turn). Your input carries the memory pages you need: <existing-memory-pages> (recall-found). You MATCH durable facts to those pages and emit POINTERS; you do not write memory yourself.

Your job over a slice is only:

① for each slice message:
     durable + matches a provided page  → emit a memoryRefs POINTER (path + note)
     durable + NO page matches          → inline into summary (degraded; extract will capture it)
     ephemeral                          → carry THIN in the digest (ages out)
     noise                              → DROP
② emit — the thin ephemeral digest + POINTERS to every matched page

Input you receive

  • slice: the raw messages, oldest first, each [msgId] @username (role): text.
  • conversationType: group | direct.
  • <existing-memory-pages>: pages found by DEVICE-LOCAL recall over this slice that may already hold durable facts from it (a fuzzy match, not an exact raw→page mapping).

Pointer-first rule (recall-fed)

You are given <existing-memory-pages> (pages recall thinks already hold durable facts from this slice). For a durable fact in the slice that matches an existing page, emit a memoryRefs pointer { path, note } — do not restate it in summary. Only inline a durable fact into summary (degraded) when no existing page matches it.

Flow

Triage each slice message:

  • Durable ("would matter in a different conversation next week?") → does a page in <existing-memory-pages> already hold it? Yes → emit a memoryRefs pointer (path + a short note), do NOT restate it in summary. No → inline it into summary (degraded); the extract hook captures it to memory separately and a later pass — once recall finds that page — will pointer it.
  • Ephemeral ("only matters to keep this thread coherent?") → thin digest line.
  • Noise ("lose it, nothing changes?") → drop.

Then compose the segment output (below): the thin ephemeral residue + the pointer list (pages you matched).

Not a re-summary: your unique work is (a) the ephemeral residue and (b) pointing durable facts at their memory pages. Durable capture (writing pages) is the extract hook's job, not yours.

Show full SKILL.md (239 more words)Show less

Speaker attribution → into the memory page's provenance

For group slices, who decided / who objected IS content. Carry attribution into the memory page (who decided, who dissented). Once it is in the page, you do NOT restate it in the digest — the pointer resolves to the page.

Output shape

Return ONLY this JSON — no prose, no fences:

json
{
  "summary": "<3-8 short lines: current thread state, referencing durable parts by memory:<path>; NOT a re-statement of the slice>",
  "salientFacts": {
    "memoryRefs":          [ { "path": "decisions/vendor.pkf", "note": "vendor B (compliance); @ceo decided, @eng dissented" } ],
    "openThreads":         [ "awaiting @ceo sign-off on A4 vs Letter page size" ],
    "abandonedDirections": [ "tried gpt-4o for compaction — too slow, dropped" ]
  }
}
  • memoryRefs — one POINTER per matched page ({ path, note }).
  • openThreads / abandonedDirections — the ephemeral residue only. Omit if empty.
  • No decisions / entities / preferences / commitments fields — those are durable, they live in memoryRefs, never inline.

Size check: the digest is the ephemeral residue + pointers, a fraction of the slice. If summary reads like a retelling, you are duplicating memory — cut to thread-state.

Failure / degradation

  • No matching pages (empty <existing-memory-pages>) → inline the durable facts into summary (degraded) with empty memoryRefs. extract still captures them to memory, and a later pass — once recall finds those pages — will pointer them. Never fail the compaction.
  • Slice is pure noise → { "summary": "(no substantive content)", "salientFacts": {} }.

Anti-patterns

  • ❌ Restating in the digest what lives in memory — point to the page.
  • ❌ Re-implementing memory triage/authoring here — delegate to the memory skill.
  • ❌ A faithful blow-by-blow summary — you produce a projection, not a transcript.
  • ❌ Dropping attribution on a decision/objection — it goes into the page's provenance.
  • ❌ Promoting ephemeral state to memory (current open question, in-flight step) — that stays in the thin digest and ages out.
  • ❌ Posting the digest as a chat message — machinery; return the structured object only.

© Prismer-AI, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in sdk/cloud/catalog/skills/conversation-compaction of Prismer-AI/PrismerCloud.

Open the folder on GitHubat commit e5d9444

Compare with similar skills

Conversation Compaction next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Conversation Compaction compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Conversation Compaction this skillPrismer-AI/PrismerCloud1.6k—~1.5kAutomated safety check: PassMIT
Modeling Conversion MetricsPostHog/posthog40k—~1.4kAutomated safety check: PassCustom licence
Strategic Compactaffaan-m/ECC276k1 repos~2.1kAutomated safety check: PassMIT
Find Project Anomaliespenpot/penpot61k—~1.2kAutomated safety check: PassMPL-2.0
Project Status Artifactanthropics/claude-plugins-official38k—~5.1kAutomated safety check: PassApache-2.0
Projects Work Managementsickn33/agentic-awesome-skills47k1 repos~4.1kAutomated safety check: PassMIT

Similar skills

  • Modeling Conversion Metrics

    PostHog/posthog

    Official

    Build reusable conversion models — funnel/step conversion rates, drop-off, and time-to-convert — on either PostHog data-warehouse views (HogQL) or an external dbt project.

    40k GitHub stars~1.4k tokensUpdated today
    Data & AnalyticsAuto-check passed

  • Strategic Compact

    affaan-m/ECC

    Suggests manual context compaction at logical intervals to preserve context through task phases rather than arbitrary auto-compaction.

    276k GitHub starsUsed in 1 repo~2.1k tokens
    Agent WorkflowsAuto-check passed

  • Find Project Anomalies

    penpot/penpot

    Check a GitHub milestone against the Main project board, report the five anomaly types to tmp/<MILESTONE-ANOMALIES.md, and fix missing milestone assignments on request.

    61k GitHub stars~1.2k tokensUpdated today
    Product & Project ManagementAuto-check passed

  • Project Status Artifact

    anthropics/claude-plugins-official

    Official

    Builds and publishes a tabbed status page for a project with several workstreams, kept current by refreshing the same private page.

    38k GitHub stars~5.1k tokensUpdated today
    Product & Project ManagementAuto-check passed

  • Projects Work Management

    sickn33/agentic-awesome-skills

    Project register: owner, team, priority, progress percentage, milestones, deliverables and budget against actual cost.

    47k GitHub starsUsed in 1 repo~4.1k tokens
    Product & Project ManagementAuto-check passed

  • Project Manager

    RightNow-AI/openfang

    Project management expert for Agile, estimation, risk management, and stakeholder communication

    18k GitHub stars~960 tokensUpdated 3 mo ago
    Product & Project ManagementAuto-check passed

More from Prismer-AI/PrismerCloud

All 88 skills in this repo

  • Prismer Google Workspace

    Prismer-AI/PrismerCloud

    Gives an agent account-scoped access to Gmail, Calendar, Drive, Contacts, Docs and Sheets through the gws CLI or a bundled Python client.

    1.6k GitHub starsUsed in 3 repos~4.2k tokens
    Auto-check passed

  • Prismer Skill Creator

    Prismer-AI/PrismerCloud

    Walks an agent through creating, importing, editing, validating, testing and publishing Prismer Skills with a fixed workflow and bundled scripts.

    1.6k GitHub stars~2.6k tokensUpdated 9 days ago
    Auto-check: notes

  • Himalaya Email CLI

    Prismer-AI/PrismerCloud

    Operates a mailbox from the terminal with the external Himalaya CLI over IMAP, SMTP, Notmuch or Sendmail, separate from any built-in email gateway adapter.

    1.6k GitHub starsUsed in 2 repos~2.3k tokens
    Auto-check passed

  • Prismer Image Generation

    Prismer-AI/PrismerCloud

    Generates one image from a text prompt with a bundled Node.js helper and delivers it once as the attachment to the current Prismer reply.

    1.6k GitHub stars~1.4k tokensUpdated 9 days ago
    Auto-check passed

  • Manim Explainer Videos

    Prismer-AI/PrismerCloud

    Produces 3Blue1Brown-style explainer animations with Manim Community Edition for math, algorithms, equations and architecture diagrams, with planning and rendering references.

    1.6k GitHub starsUsed in 2 repos~3.1k tokens
    Auto-check passed

  • Prismer Role Builder

    Prismer-AI/PrismerCloud

    Creates or updates Prismer role templates from a persona, SOP or job description, and turns a role into a working agent that runs its first task through a bundled script.

    1.6k GitHub stars~2.3k tokensUpdated 9 days ago
    Auto-check: notes

Questions about Conversation Compaction

What does Conversation Compaction do?

Produce the SESSION PROJECTION of memory for an aging conversation slice. Conversation Compaction is an agent skill from Prismer-AI/PrismerCloud. Produce the SESSION PROJECTION of memory for an aging conversation slice.

How do I install Conversation Compaction in Claude Code?

Run `npx skills add Prismer-AI/PrismerCloud --skill conversation-compaction -a claude-code`. Or copy the skill folder (sdk/cloud/catalog/skills/conversation-compaction in Prismer-AI/PrismerCloud) into .claude/skills/conversation-compaction in your project. Claude Code loads it when a task matches its description.

How do I install Conversation Compaction in Codex?

Run `npx skills add Prismer-AI/PrismerCloud --skill conversation-compaction -a codex`. Or copy the skill folder (sdk/cloud/catalog/skills/conversation-compaction in Prismer-AI/PrismerCloud) into .agents/skills/conversation-compaction in your project. Codex loads it when a task matches its description.

Can I use Conversation Compaction in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add Prismer-AI/PrismerCloud --skill conversation-compaction -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/conversation-compaction, .gemini/skills/conversation-compaction, .github/skills/conversation-compaction and .opencode/skills/conversation-compaction in your project.

What does Conversation Compaction need to run?

SKILL.md names no scripts, command-line tools or credentials: Conversation Compaction is instructions for the agent only.

Does Conversation Compaction access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Conversation Compaction safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Conversation Compaction use?

Conversation Compaction is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Conversation Compaction use?

About 1.5k tokens (SKILL.md is roughly 6.1k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Conversation Compaction?

Skills that share tags, products or a category with Conversation Compaction: Modeling Conversion Metrics (PostHog/posthog, 40k stars), Strategic Compact (affaan-m/ECC, 276k stars), Find Project Anomalies (penpot/penpot, 61k stars) and Project Status Artifact (anthropics/claude-plugins-official, 38k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Conversation Compaction?

Prismer-AI (a GitHub organization) maintains it in Prismer-AI/PrismerCloud, which has 1,555 GitHub stars. The repository holds 88 skills in this directory. The repository was last updated on September 30, 2026.

Source: Prismer-AI/PrismerCloud on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.