Home / linkerdog / rara · GitHub

context-cache-management SKILL.md skillA

context-cache-management from linkerdog/rara is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned

mdr add linkerdog/rara/context-cache-management@git:20260506.4e0a138

Pins git:20260506.4e0a138 in mdr.lock. Pin the fingerprint instead.

Audit grade: A, 17 of 17 checks passed
Kind: SKILL.md skill
Repository: linkerdog/rara
Path: .agents/skills/context-cache-management/SKILL.md
Versions recorded: 1
Latest version: git:20260506.4e0a138, committed May 6, 2026
Size: 4,602 bytes
Repository stars: 17
License: Apache-2.0

Extend RARA context compression, provider cache behavior, memory placement, or model context budgeting while preserving stable prompt prefixes.

Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.

What the file says

# Context Cache Management

Use this skill when changing RARA's context compression, tool-result projection,
provider cache handling, memory placement, context budget calculation, or
model-provider integration.

## Core Rules

- Treat the local transcript as source of truth. Request-time projection may
  shrink model input, but it must not rewrite persisted history unless the user
  explicitly asked for lifecycle compaction.
- Preserve `tool_use` / `tool_result` pairing. Do not remove a block from only
  one side of the pair.
- Keep stable prompt prefixes stable:
  1. system prompt
  2. tool schemas
  3. stable skills and project memory
  4. compacted history and carry-over
  5. retrieval and volatile recent context
  6. latest user input
- Keep retrieved memory in the volatile suffix unless it has been explicitly
  promoted to a stable workspace memory prompt source.
- Never infer cache-edit support from OpenAI-compatible request shape. Cache
  editing is a provider capability, not a protocol default.
- Do not inject provider-specific cache-retention parameters unless the backend
  explicitly declares retention control support.

## Provider Cache Capability Checklist
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint

mdr add linkerdog/rara/context-cache-management@sha256:b4d4e83c01939205

The line at the top of this page pins the label git:20260506.4e0a138, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.

Badge

[![mdr](https://markdownregistry.com/badge/art_7nwu4rhooyliuzcs.svg)](https://markdownregistry.com/a/art_7nwu4rhooyliuzcs)

1 badge image fetches in 30 days. What the badge shows.

Versions

The registry has recorded 1 version of this file, committed on May 6, 2026.

version	committed	commit	size	audit
git:20260506.4e0a138 latest	2026-05-06	4e0a138	4,602 B	A	view

Audit of the latest version

A 17 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.

pass: Frontmatter block present
pass: Frontmatter declares a name
pass: Frontmatter declares a description
pass: Size between 200 bytes and 200 KB (4602 bytes)
pass: No zero-width or bidi control characters
pass: No instruction hidden inside an HTML comment
pass: No link to an exfiltration or paste host
pass: No credential-shaped string
pass: No instruction to send local credentials anywhere
pass: No text hidden with inline styles
pass: No prompt-injection phrasing
pass: No curl or wget piped into a shell
pass: No recursive delete of root, home or parent
pass: No instruction to read or print local credentials
pass: No base64 blob over 200 characters
pass: No link to a raw IP address
pass: No script tag

Source

GitHub

linkerdog/rara · 17 stars · license Apache-2.0 · pushed 2026-10-26 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_7nwu4rhooyliuzcs
GET https://markdownregistry.com/api/v1/resolve?ref=linkerdog/rara/context-cache-management
GET https://markdownregistry.com/api/v1/blob/b4d4e83c0193920538d1da7e7f22897dc05a0c12cd7b54893dc1f72316ea733c

Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.

Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.