# Context Compressor > Compress and conserve conversation context by using terse responses, compact progress updates, explicit context pruning, compact code/content summaries, and concise handoff summaries. Use when the user asks to reduce token usage, avoid excessive context consumption, compress context, summarize current state, keep replies brief, switch to low-token mode, preserve only task-critical facts, or prepa… - **Type:** Skill - **Install:** `agentstack add skill-cyfung1031-skills-v1-2-3` - **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior - **Seller:** [cyfung1031](https://agentstack.voostack.com/s/cyfung1031) - **Installs:** 0 - **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills) - **Latest version:** 0.1.0 - **License:** MIT - **Upstream author:** [cyfung1031](https://github.com/cyfung1031) - **Source:** https://github.com/cyfung1031/skills/tree/main/context-compressor/v1.2.3 ## Install ```sh agentstack add skill-cyfung1031-skills-v1-2-3 ``` Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent. ## About # Context Compressor Use this skill to minimize context growth while preserving enough state to keep working accurately. Keep this skill's own output short. ## When to use Use this skill when any of these hold: - The user asks to reduce tokens, keep replies brief, compress context, summarize current state, preserve only task-critical facts, switch to low-token mode, or prepare a continuation/handoff summary. - A system, developer, runtime, or tool signal indicates compaction, truncation, continuation, limited context, or token pressure. - The thread is long enough that losing earlier decisions, files, tests, constraints, or non-obvious findings would cause rework or correctness risk. "Long enough" is qualitative, not token math. Use visible signs such as 10+ substantial turns, large pasted content, repeated tool output, multiple changed files, or several accepted decisions. Do not estimate exact context limits without reliable telemetry. Do not activate just because a task is mildly complex or because a simple question appears inside an old thread. Not this skill: one-off document/article summarization, one-off shortening of a single answer, or ordinary code review. Those are normal tasks unless the user also asks for a standing compact mode or continuation summary. Treat an explicit low-token request as standing until the user asks for normal/verbose/detailed mode or the task ends. Re-apply it from conversation history; the skill stores no state. Proactive activation is temporary: use it for the current phase, compaction event, or handoff, then silently return to normal when the task ends or shifts to an unrelated goal. Mention the transition only if the user asked about mode/state, silence would confuse them, or the turn ends with a handoff/summary. ## Operating mode Default to short, direct answers: 1-4 bullets or one compact paragraph unless the user asks for detail. In a long task: - State the current action and why it matters; skip restating stable instructions, settled decisions, and raw command output. - Replace verbose reasoning with decisions, assumptions, verification results, and remaining risks. - Prefer exact identifiers, such as file paths, command/test names, symbols, branches, issue numbers, dates, and source names, over prose. - When blocked, ask at most one clarifying question; otherwise make a reasonable assumption and proceed. - Record assumptions only when they affect behavior, scope, safety, correctness, permissions, or output format. - Do not claim exact token savings, remaining context window, truncation risk, test coverage, or completion unless supported by visible telemetry or validation. ## Priority when brevity conflicts Brevity never overrides correctness, safety, required workflow, or explicit user needs. Resolve in this order: 1. Safety, privacy, permissions, policy. 2. Required tool/artifact workflow, validation, citations, format constraints. 3. The user's latest explicit request for detail, full code, examples, or explanation. 4. Compression and terse style. If the user asks for both terse and detailed output, answer the concrete ask first in the most useful form. For code, start with the code or patch, then add a brief note. For a decision or diagnosis, start with the conclusion, then give only needed evidence. Use a summary-first layout only when it improves usability and does not delay the requested artifact or answer. ## Context pruning: what to keep vs. drop Keep: active goal and latest correction; constraints affecting behavior, safety, permissions, citations, or format; current plan state, blockers, decisions, and behavior-changing assumptions; files changed, commands run, test status, validation status, remaining work, and immediate next actions; non-obvious findings from code, tools, docs, files, or sources; minimal code/content structure needed to continue, such as paths, public interfaces, names, data shapes, invariants, error messages, changed logic, source titles, or line ranges. Drop or compress: rejected alternatives; full command logs once summarized; repeated explanations of standard process; large pasted content once relevant facts are extracted; large code blocks once structure, intent, interfaces, and risky details are captured; old plans superseded by the latest user correction; filler, motivational language, and redundant caveats. After compaction, continuation, or resume, treat summaries as working notes. Re-verify load-bearing facts before acting on them: file existence, current branch, changed files, failing tests, external state, citations, tool outputs, and whether the latest user request supersedes earlier work. ## Compressing code and content Do not preserve full code blocks by default. Summarize structurally unless exact code is needed next: ```markdown Path: file/repo path or source label Symbols: function/class/API names and signatures Behavior: what it does State: inputs, outputs, side effects, dependencies Issues: bugs, failing tests, stack traces, risky edges Needed: exact snippets still required, if any ``` Keep exact snippets only when small and necessary: a failing line, stack frame, regex, query, schema, migration, prompt, config key, API signature, patch-sized block to copy verbatim, or content where whitespace, ordering, quoting, or syntax is the point. For non-code content, preserve the structure and decision-critical details, not the full text: ```markdown Source: file/document/source label Topic: what this content is about Key facts: compact bullets with source anchors if available Decisions: conclusions or accepted edits Open items: unresolved questions, missing evidence, or next checks Needed: exact wording still required, if any ``` ## Handoff and update formats Full handoff: use when asked to compress, hand off, continue, or prepare a continuation. Also use it when a trigger above fires and losing state would create rework or correctness risk. Omit empty fields. `Verified` must name evidence actually checked; use `Risks` for unverified items, blockers, assumptions, or likely rework causes. ```markdown Goal: ... State: ... Decisions: ... Assumptions: ... (only if behavior-affecting) Changed: ... Verified: ... (evidence actually checked; state if partial or absent) Next: ... (next owner/action if not continuing immediately) Risks: ... (unverified items, blockers, assumptions, or likely rework causes) ``` Example: ```markdown Goal: Fix checkout tax rounding bug in `billing-service`. State: Reproduced mismatch on JPY orders; isolated to per-line rounding before discount allocation. Decisions: Round only after order-level tax aggregation; keep USD behavior unchanged. Assumptions: Tax API contract cannot change. Changed: `src/tax/calculate.ts`, `tests/tax-rounding.test.ts`. Verified: `npm test -- tax-rounding` passes; full suite not run. Next: Run full tests, then prepare patch summary. Risks: Multi-currency fixtures may hide another rounding path. ``` Use absolute dates in handoffs when relative dates could become ambiguous. Ordinary update: use only fields carrying new information; do not pad for rhythm. Add `Verified` only when evidence was checked. ```markdown Now: ... (current action or status) Found: ... (new discovery, result, blocker, or verification outcome) Verified: ... (evidence actually checked) Next: ... (only if the next step changed or would otherwise be unclear) ``` ## Response rules Answer the latest request first: requested code, artifact, decision, or summary goes before background. Use dense but readable wording, summaries over transcripts, and exact dates, paths, branches, issue numbers, test names, source names, and citations when relevant. State uncertainty in one sentence instead of hiding it. Expand only when accuracy, safety, citations, validation, required workflow, or the user's explicit request demands it. Completion claims need evidence. Report what was checked; if validation was partial or not performed, state what remains unverified. ## Interaction with other skills When another skill is also required, follow that skill's substantive workflow. Use this skill only to reduce narration, repeated setup, transcripts, and nonessential explanation. Do not skip required validation, citations, safety checks, artifacts, tool steps, or user-requested detail just to save tokens. ## Source & license This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code. - **Author:** [cyfung1031](https://github.com/cyfung1031) - **Source:** [cyfung1031/skills](https://github.com/cyfung1031/skills) - **License:** MIT Install and usage instructions live in the source repository linked above. ## Pricing - **Free** — Free ## Security capabilities Automated source analysis of v0.1.0 — what this tool can access: - **Network access:** no - **Filesystem access:** no - **Shell / process execution:** no - **Environment & secrets:** no - **Dynamic code execution:** no *"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.* ## Versions - **0.1.0** — security scan: passed — Imported from the upstream source. ## Links - Listing page: https://agentstack.voostack.com/l/skill-cyfung1031-skills-v1-2-3 - Seller: https://agentstack.voostack.com/s/cyfung1031 - Browse the marketplace: https://agentstack.voostack.com/browse --- Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.