dsh-funnel
在工具输出进入模型上下文之前进行整理：保留错误/警告行以及开头和结尾内容，将完整文本写入磁盘并提供指针。加快交互、缩小上下文、提升注意力——适用于所有工具，而不仅仅是 bash。
插件会安装到这里；不确定时保持 web。
npx -y @deepseek-ai/dsh plugin --profile web add dsh-funnel@0.1.0Funnel 以 dsh-funnel 发布，当前版本为 0.1.0。Plugin Hub 会校验它的 manifest，并保存精确安装来源，便于复现安装结果。
English | 中文
Curate tool output before it enters the model's context — keep the lines that matter, spill the rest to disk with a pointer. Faster turns, smaller context, sharper attention, for every tool.
Every tool result is appended to the conversation and re-sent on every subsequent model request. A npm test run that prints 500 lines rides along for the rest of the session: you pay for it on each turn, it fills the context window, and it buries the one AssertionError the model actually needs.
The existing layers don't cover this:
| Layer | What exists | What it doesn't do |
|---|---|---|
| Capture (executor) | 64 MB spill protects the process from OOM | nothing about what the model sees |
| Execution (bash only) | maxOutputBytes byte-cap, head truncation | shell tools only; blind cut, no error lines kept |
| Compaction time | official tool-result-pruner rewrites over-budget results on the surface | fires only when compacting; blind head/tail; model can't read the original back |
| Ingestion — every result, immediately | — | ← dsh-funnel |
Compaction-time pruning cleans up after the budget is already blown; dsh-funnel keeps it from being blown. They compose.
Install and forget. One command, zero configuration. The plugin covers every tool that returns text — bash, file reads, web fetch, search — and results under the threshold pass through byte-identical, so you never notice it until it saves you.
A ~24× smaller model view, measured. In a real headless session, a 2001-line (8,893-byte) bash result reached the model as 370 characters: head, tail, and every error/warning line kept:
- [2001 lines of raw output]
+ [dsh-funnel] output curated: 2001 lines → 40 kept (head 15, tail 25, 0 pattern-matched). Full output: .dsh-funnel/2026-08-15T09-21-56-548Z-bash-1.log
+ 1 … 15
+ [dsh-funnel] … 1959 similar lines omitted …
+ 1977 … 2000
Every later turn in the session pays for 370 characters instead of 8,893 — the saving compounds each request.
Nothing is lost, and the model recovers on its own. The full text spills to .dsh-funnel/ and the notice carries the path. In testing, the model missed a middle line in the curated view, searched the spill file itself, and answered correctly — curation is pagination, not deletion. The spool doubles as an audit trail of what the agent actually saw.
Long tasks hold up better. Smaller context means cheaper turns, later compaction, and less attention dilution — the agent stays sharp deeper into a session instead of degrading as junk accumulates.
Tune it to your workflow. Threshold, kept lines, keyword patterns (add your project's own failure markers), per-tool aggressiveness, spill on/off.
Boundaries: error results are never touched (they are short and load-bearing); this complements compaction rather than replacing it.
dsh-funnel wraps the tools/execute extension point (the seam the official timeout guard uses) and curates the canonical value of successful results: the registry re-projects model-facing content from value, so the curated view is exactly what the model sees and what the session log records. Long string fields inside value — stdout, stderr, fetched pages, file contents — are replaced in place; error results are never touched.
dsh plugin add github:YuanyuanMa03/dsh-funnel
Works on every tool that returns text — bash, file reads, web fetch, search.
All fields optional; set them under the plugin entry in your profile:
- id: dsh-funnel
name: 'dsh-funnel'
config:
maxChars: 4000 # results at or below this length pass through untouched
keepHead: 15 # lines kept verbatim from the head
keepTail: 25 # lines kept verbatim from the tail (errors usually live here)
keepPatterns: [error, warn, fail, fatal, exception, assert, denied, panic]
maxLineChars: 1000 # single oversized lines (minified/JSON) are capped
spill: true # write the full text to disk and reference it
tools: # per-tool overrides
web_fetch: { maxChars: 2000 }
maxChars return as the same object; zero allocation when idle.v0.1.0 — verified end-to-end in a real headless session: a 2001-line bash result was curated to 40 lines in model view and log, and the model recovered the omitted middle by reading the spill file (its own words: "the tool curates long output for display … verified against the complete saved output"). Unit tests cover the curation logic and the plugin wrapper. Planned: session-level savings accounting ("this session: N tokens avoided").