context-budget SKILL.md skillA

context-budget from ericrisco/rsc-harness is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add ericrisco/rsc-harness/context-budget@git:20260729.3ecb845

Pins git:20260729.3ecb845 in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 17 of 17 checks passed
Kind
SKILL.md skill
Repository
ericrisco/rsc-harness
Path
skills/context-budget/SKILL.md
Versions recorded
2
Latest version
git:20260729.3ecb845, committed Jul 29, 2026
First version recorded
Jun 2, 2026
Size
9,605 bytes
Repository stars
236
License
MIT

Use when a long-horizon task is filling the context window and you must decide what to keep, offload, drop, or hand off to a fresh window — when to compact, what the summary must preserve, and whether to isolate a read-heavy subtask in a subagent. NOT dollar spend or caps (that is cost-tracking), NOT finding context via embeddings (that is rag).

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 budget

The context window is RAM, not a hard drive. Full ≠ free: a window stuffed to 95% does not just cost more money, it *reasons worse*. Model performance degrades as input tokens grow — even well inside the stated limit, every token added depletes a finite attention budget (Chroma "Context Rot" research, accessed 2026-06-02). Your job on a long task is to keep the *live* window lean and externalise everything else, so the work can run for hours across many fresh windows without losing the thread.

> **The one rule:** if you can reconstruct a thing from a file or from git, it does not belong resident in the window. Keep load-bearing-right-now; evict the rest. The cost of forgetting is one re-read; the cost of hoarding is silent quality rot on every turn that follows.
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add ericrisco/rsc-harness/context-budget@sha256:aaceab7844921305

The line at the top of this page pins the label git:20260729.3ecb845, 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 badge

[![mdr](https://markdownregistry.com/badge/art_ftamjhy3ekfwkajd.svg)](https://markdownregistry.com/a/art_ftamjhy3ekfwkajd)

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

Versions

The registry has recorded 2 versions of this file, between Jun 2, 2026 and Jul 29, 2026, about one every 57 days.

versioncommittedcommitsizeaudit
git:20260729.3ecb845 latest2026-07-29 3ecb845 9,605 BA view · diff
git:20260602.99b4be12026-06-02 99b4be1 11,140 BA 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 (9605 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

ericrisco/rsc-harness · 236 stars · license MIT · pushed 2026-10-21 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_ftamjhy3ekfwkajd
GET https://markdownregistry.com/api/v1/resolve?ref=ericrisco/rsc-harness/context-budget
GET https://markdownregistry.com/api/v1/blob/aaceab78449213053f34af02071f0a7620aa2035b068954566c361dd15fd8fc5

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.

More from ericrisco/rsc-harness

llms.txt@site llms.txt file
ericrisco/rsc-harness · site/llms.txt
git:20261021.031a776 · audit A · 236 stars
ab-testing SKILL.md skill
ericrisco/rsc-harness · skills/ab-testing/SKILL.md · Use when designing or analyzing a controlled experiment — falsifiable hypothesis, sample size from an MDE, reading…
git:20260729.3ecb845 · audit A · 236 stars
accessibility SKILL.md skill
ericrisco/rsc-harness · skills/accessibility/SKILL.md · Use when making a web UI conform to WCAG 2.2 Level AA — axe-core or Lighthouse a11y violations, keyboard operability…
git:20260729.3ecb845 · audit A · 236 stars
ads SKILL.md skill
ericrisco/rsc-harness · skills/ads/SKILL.md · Use when running or fixing paid acquisition on Google or Meta — campaign structure (Performance Max, Demand Gen…
git:20261011.c7821cd · audit A · 236 stars
agent-eval SKILL.md skill
ericrisco/rsc-harness · skills/agent-eval/SKILL.md · Use when measuring whether an LLM or agent system actually got better and gating merges on it: golden sets, fixing an…
git:20260729.3ecb845 · audit A · 236 stars
agent-safety SKILL.md skill
ericrisco/rsc-harness · skills/agent-safety/SKILL.md · Use when bounding an LLM agent that already runs — scoping its task domain, gating tools to least privilege, defending…
git:20260729.3ecb845 · audit A · 236 stars
ai-media SKILL.md skill
ericrisco/rsc-harness · skills/ai-media/SKILL.md · Use when a creative goal must become a finished media file: pick and order generative-media models per modality — AI…
git:20261021.031a776 · audit A · 236 stars
analytics SKILL.md skill
ericrisco/rsc-harness · skills/analytics/SKILL.md · Use when instrumenting product or web analytics — GA4/PostHog SDK wiring, event taxonomy, funnels, double-counted…
git:20260729.3ecb845 · audit B · 236 stars
analyze SKILL.md skill
ericrisco/rsc-harness · skills/analyze/SKILL.md · Use when constitution, spec, plan and tasks all exist and you want them cross-read against each other before any code…
git:20261003.f0c84d6 · audit A · 236 stars
angular SKILL.md skill
ericrisco/rsc-harness · skills/angular/SKILL.md · Use when building, refactoring, or debugging Angular (v20/21+): standalone components, signals, zoneless change…
git:20260729.3ecb845 · audit A · 236 stars
api-connector-builder SKILL.md skill
ericrisco/rsc-harness · skills/api-connector-builder/SKILL.md · Use when writing a client for someone else's REST or GraphQL API: auth flow choice and token refresh, pagination to…
git:20260729.3ecb845 · audit A · 236 stars
api-design SKILL.md skill
ericrisco/rsc-harness · skills/api-design/SKILL.md · Use when settling the contract of an API you expose, before implementation: resources/URLs, REST vs GraphQL…
git:20260729.3ecb845 · audit A · 236 stars

Every file in ericrisco/rsc-harness

Other files named context-budget

context-budget SKILL.md skill
affaan-m/ecc · docs/ja-JP/skills/context-budget/SKILL.md · エージェント、スキル、MCPサーバー、ルールにわたってClaude Codeのコンテキストウィンドウ消費を監査します。肥大化、冗長なコンポーネントを特定し、優先順位付けされたトークン節約の推奨事項を生成します。
git:20260517.ec9ace9 · audit A · 284,510 stars
context-budget SKILL.md skill
coco-research/coco · skills/context-budget/SKILL.md · Audit token consumption across agents, skills, MCP servers, and rules. Identifies bloat, redundant components, and…
v0.1.0 · audit A · 827 stars
context-budget SKILL.md skill
onewave-ai/claude-skills · context-budget/SKILL.md · Use when a coding session is burning context fast, output is drowning in noise, the model keeps forgetting earlier…
git:20260920.97b5147 · audit A · 358 stars
context-budget SKILL.md skill
frankxai/claude-skills-library · absorbed/harness/context-budget/SKILL.md · Count what a session loads before adding another skill or MCP server. Use when the default prompt is heavy, a new…
git:20261010.682b2ce · audit A · 47 stars
context-budget SKILL.md skill
felvieira/claude-skills-fv · skills/49-context-budget/SKILL.md · Audita o peso de contexto efetivamente observável na sessão — instruções raiz, hooks, MCP anunciados, rules ativas e…
v1.0.0 · audit A · 24 stars
context-budget SKILL.md skill
univers42/claude-deal-with-the-devil · skills/context-budget/SKILL.md · Measure and cut what a config costs in context every session: always-on rules, invocable descriptions, skills that…
git:20261017.fcaf2cc · audit A · 4 stars

Browse by kind, by grade A, or by owner.