Plugins

Filters tool outputs, prunes stale context and discloses skills on demand for DSH 0.2.1-alpha.1.

AST-based context compression, test log filtering, and token budget guard for DeepSeek Harness: strips redundant syntax, prunes noisy log lines, and enforces a token ceiling so the context window never overflows.

Full-type session lifecycle management: one-shot subagents archived on completion, idle continuable subagents / main sessions archived (recoverable), a capacity cap, projection-cache cleanup, and a hot-reload settings panel.

DeepSeek Harness plugin: restart without losing work. It waits until every conversation is idle, restarts the host, and picks the interrupted work back up instead of dying with it; where a supervisor already manages the host it asks that supervisor, and it never starts a second host next to a running one. It also fixes a common one: stale login cookies pile up on every launch until the request header overflows and the host will not start; the plugin clears them before each restart. Every restart is preceded by a snapshot of the configuration and the installed plugins - 2 are kept by default, up to 10 are allowed, and old or incomplete ones are pruned. That snapshot is never applied automatically: the plugin does not watch the host and does not restore anything on its own; it is only there for you to look at, or to roll back by hand. When the work is done it counts down and powers the machine off, so nobody has to stay up at night. It works on the Windows desktop client, and on a hosted dsh web instance - verified on WSL with systemd; Windows-hosted web, WSL without systemd, and native Linux are untested.

Prunes the tool schema to a whitelist and injects a structural preamble on a session's first turn only, then passes every later turn through untouched.

Deletes the write-only streaming fragment rows from session logs on disk so a long session loads into fewer objects; the pruned log is checked against the loader contract before the file is replaced, and the previous file is kept as .orig.

Precomputes the compaction summary in the background and commits the held result when pressure compaction fires, removing the foreground model round-trip. Registers itself as the session compaction engine, pins the span it summarised so the payload fingerprint still matches at commit time, prunes tool results before committing, and falls back to the stock foreground path on payload drift or context overflow.

Loop guard for DeepSeek Harness: detects and stops repeated output mid-stream, prunes and compacts context, then automatically resumes the task without a restart.

Persona and memory from a single AGENT.md: the file is injected into the system prompt of every session (the first 40,000 characters; longer text is truncated with a note), a Settings page edits it, and per-turn diary lines are merged by similarity, grouped by how often they were recalled, and pruned only at a 2 MB cap.

Saves tokens in DeepSeek Harness sessions by reading any file by line range or symbol mention instead of whole, and by bounding oversized tool results while keeping a recovery path. Pruning happens at the source, not after content reaches the context. Measured 43.9% fewer prompt tokens in a modelled 30-step session.

Session Steward for DSH: browse and bulk-prune the official session archive set, and run a four-gate health checkup (log integrity, projection cache, lossless JSON, cold read) that ends in a reversible, backup-first prescription. Never rewrites session logs.

Manage Git worktrees inside DSH: scan workspace repositories, create worktrees, add them as DSH workspaces, open folders, remove worktrees, and prune stale records.

Blocks the end of a turn while temporary files from that turn are still lying around: it registers a listener on the serial agent/turn-stopping hook and, if anything is left, steers the model with the file list and two ways out — delete them, or justify each one and move it somewhere deliberate. Candidates come from two narrow temp-file sources, plus a backup-retention channel that is on by default and flags excess .bak copies of files touched this turn. The first is the .agent-tmp/ convention directory under the session cwd: whatever is left in it is reported outright, because that directory exists only for throwaway files. The second is the paths this turn actually wrote, captured through the fs/write-intent hook; those are reported only when they sit under that cwd, fall outside .git, node_modules, .tmp-sessions and out, match a temp-file pattern, and the file is still present — which is what keeps junk already sitting in the workspace from being resurrected as a complaint about this turn. One steer per turn, subagent sessions skipped, nothing deleted by the plugin in the default steer mode (only the opt-in mode: prune deletes backups), and the check is readdir plus stat only.

Layered context management: at each turn end older tool output becomes a deterministic card, the model can fold older ranges into its own summaries in up to three tiers (compress), and every original is archived locally for decompress and search_context. Works alongside DSH-Mem0: superseded runtime-context snapshots (such as recalled memories) are never archived, carded or folded, and are replaced by a one-line placeholder at turn end when the prompt cache is already broken, when the saving outweighs the cache miss, or above 75% of the window; a strataBudget service (pressure, allowance, countTokens) and a strata/fold-landed event are provided for other context plugins; strata_status warns when DSH's automatic compaction is off or its official summary keeps failing. Its L1 layer, archive rotation, prune guard and preset takeover are forked from dsh-live-compact 1.0.6 (MIT); uninstall dsh-live-compact or billion-context first — they replace the same compaction rows and Strata will not take over while either is installed.

Compaction ledger for the DeepSeek Harness: audit every context compaction (what was shadowed, when, at what token cost) and let the model recall pruned content on demand

Audit what context compaction actually lost. Reads the raw session event log and checks session-constraint survival (the standing instructions the user gave — do-not-X, must-Y, only-in-Z — and whether the summary still carries them; arXiv 2608.11242 measured ~17% survival across compactors), summary hallucination (paths, versions, ports and identifiers the summary invents that were never in the history it replaced), ghost references ("see above" pointing at removed text), token accounting (a summary nearly as large as what it claims to free), and unaccounted prunes. Four agent tools including a cross-session sweep that ranks conversations by how much compaction hurt them. Deterministic — no LLM, so a wrong answer is a rule bug with a name.

Durable cross-session memory plus a fetch ladder that survives hostname-keyed blocking. Four tools: memory_recall, memory_save and memory_stats over a plain-JSON fact store at $DSH_HOME/memory/facts.json, plus resilient_fetch which tries plain fetch, then a same-service alias host (Wikipedia only for now), then curl over IPv6 and IPv4 (needs curl on PATH). No daemon, no npm dependency, no network at import time. (1.3.0 shipped a fifth tool, memory_prune; it was withdrawn in 1.4.0 because adding a tool to a running plugin breaks sessions that were opened before it.)

Requires mise installed on the host. A mise-backed toolchain manager for the host machine — a settings page listing the active, installed and available versions, with .mise.toml-driven install, use, uninstall and prune, mise shims prepended to the host PATH on load, and nine agent tools for the same work (status, versions, PATH scan, install, use, prune, uninstall, activate, progress).

System One decision runtime for dsh. Ships fast non-generative tools (choice, score, screen, gate) on a local or hosted System One gateway, and — without the model calling anything — prunes oversized tool output segment by segment, refuses repeated identical calls once System One confirms stagnation, verifies completion claims against real tool evidence, and corrects failing retries. Published on npm as dsh-harness-reflex (plugin) and harness-reflex (core, also a stdio MCP server).

Local file recall: indexes the folders you choose into one local SQLite database, so 'where did I put that file?' is one query instead of a disk walk. Caches, VCS and program folders are pruned per directory; each folder keeps its own incremental anchor, so a folder added later is backfilled in full. Read-only: it never modifies, moves or deletes your files. Works on DeepSeek Harness 0.1.7 through 0.2.x, tested on 0.2.0-rc.1.

DeepSeek Harness plugin bundle: grill_tree, a fixed-policy design-space search tool. Expands one decision per node, fans out independent proposers, has a separate judge prune the branches, beam-limits each depth, and ends with complete designs or an infea

Session context triage for long DeepSeek Harness sessions: deterministic screeners trim stale, duplicate, failed, and oversized content through the official ctx.compaction seam.