hive.context-preservation
This Claude Code skill manages information retention within finite context windows by directing users to proactively extract and store critical values from tool results into working notes before automatic pruning removes them. Use this skill when working with multiple tool calls, long workflows, or tasks requiring handoffs, where you need to preserve specific URLs, API fields, IDs, or analysis conclusions that you'll reference later in the conversation.
git clone --depth 1 https://github.com/aden-hive/hive /tmp/hive.context-preservation && cp -r /tmp/hive.context-preservation/core/framework/skills/_default_skills/context-preservation ~/.claude/skills/hive.context-preservationSKILL.md
## Operational Protocol: Context Preservation
You operate under a finite context window. Older tool results WILL be pruned. Extract what you need while it's still in context.
**Save-as-you-go.** After any tool call producing information you'll need later, immediately extract the key data into `_working_notes` or `_preserved_data`. Do not rely on referring back to old tool results — once they're pruned they're gone.
**What to extract:**
- URLs and key snippets (not full pages)
- Relevant API fields (not raw JSON blobs)
- Specific lines, values, or IDs (not entire files)
- Analysis conclusions (not raw data)
**Handoffs between tasks** happen through `progress.db`, not through shared-buffer handoff blobs. When you finish a task, any state the next worker needs goes into the task row itself (`steps.evidence`, `tasks.last_error`, `sop_checklist.note`) — see `hive.colony-progress-tracker`. Use `_working_notes` for things the DB schema doesn't cover.
You will receive an alert when context reaches {{warn_at_usage_ratio_pct}}% — preserve immediately.SOP for debugging browser automation failures on complex websites. Use when browser tools fail on specific sites like LinkedIn, Twitter/X, SPAs, or sites with Shadow DOM.
Required before any hive-browser CLI command. The browser is driven from the terminal by running `hive-browser <command> ... --json` via terminal_exec — not via MCP tools. Teaches the browser lifecycle rules (the bridge attaches to the USER'S running Chrome — never kill or launch browser processes; timeouts are transport issues, not crashes), the screenshot + coordinate workflow (hive-browser interact with a fractional coordinate) that reaches shadow-DOM inputs selectors can't see, the viewport-fraction coordinate rule (not pixels), rich-text editor quirks ("send button stays disabled" failures), and CSP gotchas. Covers Chrome via CDP through the GCU Beeline extension.
Required reading whenever any chart_* tool is available. Teaches the one-tool embedding contract (call chart_render → live chart appears in chat AND a downloadable PNG lands in the queen session dir), the ECharts (data viz) vs Mermaid (structural diagrams) decision, the BI/financial-grade aesthetic baseline (no chartjunk, restrained palette, proper typography, single message per chart), and the canonical spec patterns for the 12 most-common chart types. Skipping this leads to 1990s-Excel charts, missing downloads, and the agent writing markdown image links by hand instead of letting chart_render drive the UI.
Claim tasks, record step progress, and verify SOP gates in the colony SQLite queue. Applies when your spawn message includes a db_path field.
Follow a structured recovery decision tree when tool calls fail instead of blindly retrying or giving up.
Required before calling image_generate. Create and edit images from a prompt — generate an image, make a picture / logo / illustration / icon / banner / poster / thumbnail / hero image / mockup / product shot / social graphic, or edit / restyle / combine existing images from reference images. Uses OpenAI gpt-image-2 through the Hive image service, billed to the user's Hive credits like an LLM call (no API key needed). Teaches the exact call shape, the quality/cost tradeoff (quality="low" is the default and cheapest), reference-image editing, how to show the result to the user with attach_file, and the failure modes (out of credits, model unavailable, moderation).
Read before automating LinkedIn with browser_* tools. LinkedIn combines shadow DOM (#interop-outlet), strict Trusted Types CSP that silently drops innerHTML, Lexical composer, native beforeunload dialogs that hang the bridge, and aggressive spam filters — each has bitten us at least once. Verified flows for profile messaging, connection-request acceptance, feed composition, and search. Requires hive.browser-automation. Verified against logged-in production 2026-04-11.
Maintain a free-form scratchpad of decisions, extracted values, and open questions so context pruning doesn't lose anything you still need.