context-weight-gate
Skill3 starsChanged 4 days agoMITView on GitHub
Refuse an unanchored timeline, a second timeline database, a raw screen-tree payload, a model call, and an Electron or generated native window, and keep latency and memory figures unmeasured.
In this repository153 other files it gives its agents.
- AGENTS.mdAGENTS.md
- hermes-mobile/AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- .github/copilot-instructions.mdCopilot instructions
- .cursor/rules/agent-automations.mdcCursor rule
- .cursor/rules/agent-device.mdcCursor rule
- .cursor/rules/agent-harness-router.mdcCursor rule
- .cursor/rules/agent-skills-overview.mdcCursor rule
- .cursor/rules/always-agent-mode.mdcCursor rule
- .cursor/rules/callstack-agent-skills-overview.mdcCursor rule
- .cursor/rules/callstack-react-native-best-practices.mdcCursor rule
- .cursor/rules/codeql-agent-hygiene.mdcCursor rule
- .cursor/rules/decision-stack.mdcCursor rule
- .cursor/rules/expo-credit-budget.mdcCursor rule
- .cursor/rules/fleet-repo-intelligence.mdcCursor rule
- .cursor/rules/hermes-mobile-prevention-july-2026.mdcCursor rule
- .cursor/rules/hermes-mobile-real-users.mdcCursor rule
- .cursor/rules/hermes-mobile-verification.mdcCursor rule
- .cursor/rules/no-desktop-hijack.mdcCursor rule
- .cursor/rules/no-manual-handoffs.mdcCursor rule
- .cursor/rules/obsidian-teammate-collaboration.mdcCursor rule
- .cursor/rules/prefer-headless-background-ops.mdcCursor rule
- .cursor/rules/react-native-best-practices.mdcCursor rule
- public/llms.txtllms.txt
- advertise-thumbgate-github-marketplaceSkill
- agent-harness-routerSkill
- agentic-pattern-selectorSkill
- agents-judgment-calls-gateSkill
- ai-native-sdlcSkill
- android-dev-summit-26-gateSkill
- anti-babysitting-mandateSkill
- anti-babysittingSkill
- background-agent-hold-gateSkill
- browseros-agent-harnessSkill
- webmcp-agent-readinessSkill
- ci-first-failSkill
- ci-pr-monitorSkill
- cloudflare-developer-stackSkill
- cloudflare-kitesurf-browserSkill
- cloudflare-monetization-gatewaySkill
- confidence-offer-gateSkill
- content-laneSkill
- context-six-blockSkill
- context-vaultSkill
- dual-model-agentic-graphragSkill
- egm-fresh-reviewSkill
- elevenlabs-voice-agentsSkill
- emeans-quant-and-instinctSkill
- explainx-trending-honestSkill
- explainx-trending-ragSkill
- future-agi-self-healing-engineSkill
- garryslist-opportunity-trackerSkill
- gemini-decisions-hold-gateSkill
- git-at-scaleSkill
- gitbutler-fleet-automationsSkill
- gitbutler-google-ssoSkill
- gitbutler-mcp-isolatedSkill
- gitbutler-session-absorbSkill
- git-hooks-no-huskySkill
- github-marketplace-action-publishSkill
- The first 60 shown.
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 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.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 8
Use this before a meeting note is treated as one clock, before screen text is sent onward, before a task is sent to a model, or before a …
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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.
Files
Context weight gate
Use this before a meeting note is treated as one clock, before screen text is sent onward, before a task is sent to a model, or before a local screen is generated from a table.
The ideas taken from https://bugsee.com/, https://www.firecrawl.dev/, https://openrouter.ai/, and https://www.jetadmin.io/ are a shared clock, a dense text extract, a cheaper path for repetitive work, and a screen built from a schema. This gate keeps the local piece and refuses the rest. It does not install those products, open a database, call a model, capture the screen, or start a window. Their latency and size claims stay theirs. Local fields stay null.
A note that only cites those pages stays clear. Leaving the menu-bar app alone stays clear. Naming a model, without a route or a call, stays clear.
What this gate does
meetingorrecord, ordayglass_get_live_meeting, without the same integer stamp on the spoken line, the terminal line, and the screen line, is blocked withunanchored_timelineandhumanReviewtrue.timestampMs: Date.now()is blocked the same way. When the three stamps match, the gate returns markdown and still leavespublishedfalse.database, or a timeline table, a virtual table, ornode:sqlite, is blocked withsecond_database. The gate does not open a database.- A command that runs
bugsee-firecrawl-openrouter-jetadmin-engineis blocked withvendor_engine. compileturns a code window into a fenced block and slide lines into headings. An accessibility tree is blocked withgeometric_noise.localCompressionPctstays null.routeorcall, a model hostname path, or a send toward a named model is blocked withremote_call. Repetitive formatting, basic tests, and schema mapping are labeledlight. A multi-file migration or abstract logic is labeledheavy. Sending the light label to a heavy model, or the heavy label to a light model, istier_mismatch.routedstays false.ui,webview,swift, or a write onto a.swiftfile is blocked withvendor_ui. A column list becomes a markdown table.processStartedandfileWrittenstay false.- Install of those products or Electron is
vendor_install. A recorded compression ratio, a recorded 93 MB idle figure, or an unattributed exact-clock, density, or cache claim isunmeasured_figure. appliedstays false.millisecondExact,localLatencyMs,localCompressionPct, andidleRamMbstay null.
Commands
node tools/context-weight-gate.js --action note --note "Leave the menu-bar app alone."
node tools/context-weight-gate.js --action route --task "formatting data"
node tests/test-context-weight-gate.jsA clear review exits 0. A blocked review exits 2.
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
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.

