Plugins

Daily session digest that grows a user profile and a companion personality, gated by evidence, and injects both into every prompt. Runs daily at 17:00 by default over up to 20 sessions per run on the host model (deepseek-v4-pro), applies the results automatically, and injects up to 1200 characters of profile and persona into every prompt. Its package name, dsh-persona-memory, is also used by other listed entries, so install only one of them.

Inspect plugins and Skills in the running DSH profile, preview updates, record verifiable snapshots and offer recovery only when replay evidence is available. Requires Node 22+ and DSH 0.2.2-rc.1 or a later declared-compatible host. Restoring a plugin version does not restore configuration, its complete dependency tree or agent presets.

A DeepSeek Harness plugin bundle: eleven specialist roles usable as one-shot consults or as team members, a team plane with a plan, a shared board and a dependency graph, a durable workmate library, a memory engine, and a verification plane that refuses a PASS unless a different agent records the verdict against gate evidence. Installs as one package with a dsh.bundle patch; also ships a native dshGo panel interface and a conformance gate pinned to a dsh-Go protocol v1 spec commit (d5ad25c).

Registers a battle-tested development delivery rulebook as a skill family - one core skill (clarification, delivery checklist, navigation) plus nine focused skills (layered architecture, evidence-driven debugging, minimal changes, impact-graded verification, migration feature parity, async UI feedback, sizing rules, CLI safety and cleanup, project notes). Only the skills matching the current task are loaded, trivial tasks (small reversible single-file edits) are exempted outright, so each session pays tokens just for the rules it actually uses.

Zero-dependency browser hand & eye over CDP: drives your own Chrome/Edge with real clicks, Chinese IME typing, scrolling and text/selector element picking, and captures whole-page evidence into one replayable folder — DOM, accessibility tree, screenshots, cookies, storage, console, HAR, resource waterfall and perf metrics.

Gives one conversation a pinned as-of point and a dated evidence window on both sides — \[as_of − before, as_of + after], past, future or BC. In restricted mode the agent is told the active window once per change, web_search results are filtered to dated sources inside the window and undated sources are rejected, and the first reply must echo the window (a missing echo gets one nudge). It also narrates long silent background jobs with deterministic \[TIME FACT] facts and exposes a read-only time_pin_status tool; default mode injects nothing and filters nothing.

Requirement-to-delivery workbench: paste or upload a requirement, and a plan-dispatch-verify loop delivers it with child agents and test evidence.

The judgment layer for DeepSeek Harness plugins: attribute why a plugin silently disappeared, judge a candidate against your exact environment, and keep a capability ledger. Offline, read-only, evidence-first.

Read-only knowledge gateway over a local Markdown vault, in four tools: `wiki_recall` returns the matching section in full with page-level citations plus vault health checks and a page skeleton for writing a good answer back; `wiki_context` returns one evidence pack with per-page excerpts, a support verdict and an explicit list of gaps; `wiki_verify` re-checks a drafted answer against that pack and reports claims the evidence cannot back, including unsupported numbers; `wiki_observe` reports how often answers needed correcting. The settings page and sidebar panel share the same query path as the model, so a search you run by hand cannot disagree with what the model was given. The three knowledge-gateway tools (`wiki_context`, `wiki_verify`, `wiki_observe`) need a local Python 3 bridge — `bridge.python` plus `bridge.toolsRoot` containing `wiki_bridge.py`; without it they return `E_PYTHON_UNAVAILABLE`.

Unified long-term memory hub: captures turns into L0 evidence, distills in the background into L1 facts / L2 scenes / L3 profile (depth configurable), recalls via hybrid BM25+fuzzy search, ships 13 memory_* tools, traceable lineage back to source turns, optional Hindsight backend, and a built-in memory panel with three entry points.

Long-conversation memory for DeepSeek Harness (DSH): keeps settled decisions, user corrections and exact numbers, with their conditions and sources, available across compaction and restarts in 100-1000+ turn sessions; one small budgeted evidence block per model step, with enabled skill rules re-applied in the same block.

Adds handoff, context-decision and context-diagnose skills, self-contained context_pressure/context_history tools (peak tokens, compaction stats), per-route realWindows with windowSource evidence, routeKey, handoffAt forced-handoff line, and slash commands.

One prompt row carrying two independently switchable halves. Identity: a fixed self-reference restated at the head, at the tail after every other section, and in the runtime snapshot, with an instruction to treat contradicting claims as stale cache. Discipline: ten rules, each paired with the condition a reviewer checks — external evidence before claiming done, change face equal to review face, correctness from named call points rather than timing, one source of truth, extension through the registered seam, runtime validation only at real boundaries, loud failure instead of silent skips, orthogonal outcomes reported separately, tests through the delivered entry point, and answer shape (conclusion first, length set by the change).

Study companion for your materials with guided practice, evidence-grounded feedback, Leitner spaced review, 18 teacher tools, and a Chinese/English student panel using the current DSH session model; requires DSH >=0.2.1-rc.2 <0.3.0-0, Python 3.11+, and Python runtime dependencies.

Forensic analysis of past model-routing decisions. Replays each session, classifies every turn's task complexity from its request shape (deterministic, no LLM), and reports two kinds of waste: an expensive model on cheap work (the difference is measurable), and a cheap model on expensive work that failed on its own merits — where every retry re-sends the whole context, so the downgrade cost more than the upgrade. Distinguishes provider-capacity retries (RATE_LIMIT/TIMEOUT/TRANSPORT) from model failures so it does not cry wolf. Derives routing rules from the evidence, priced against a built-in per-million table.

Automatic second-opinion review: a restricted child session re-reads the agent's recent work with read-only tools and delivers at most one evidence-backed note, or stays silent. Requires a model route in the config; the plugin stays disabled until one is set.

Runs a second agent: hands independent subtasks to a local Codex CLI in parallel (up to three at a time), and asks it to review a design or a plan before the user sees it, returning conclusion, evidence, counterexamples and a verification step. A dispatch missing its timeout or its acceptance check is rejected by the plugin, not by the model remembering to check.

Agent preset for DeepSeek Harness: the standard tool stack plus four working rules written into the persona prompt — a project keeps at most four documents (intro, decisions, change log, current state and next steps), those documents describe only the current state with no AI edit history or superseded decisions, no git commit or push or pull request until the user explicitly asks, and a careful-coding workflow (restate the goal and done-criteria first, surgical diffs, evidence before claiming success) with plain-language reports.

Requires the Codex CLI installed and `codex login` completed. Consult the OpenAI Codex CLI from DeepSeek Harness with three tools: codex_consult runs Codex in a read-only sandbox, inlines evidence files, verifies the served model and reasoning effort from the Codex rollout file, retries disconnected streams, and records quota before and after the call; codex_quota reports the current 5-hour and weekly quota; codex_calls lists recent calls.

Tide Memory: a long-term memory that forgets on purpose. It keeps what matters and lets the rest fade. Whether an entry enters the context is decided by an evidence score C = s/(s+2): below the threshold (0.35 by default) it never enters. Events decay by half-life (30 days, or 7 days when they carry emotion); facts never decay; superseded entries are kept and marked stale instead of deleted. Recalled lines do enter the context and therefore do consume tokens — small, but real. No embeddings, no network calls, no cloud.

Assembly-aware strawberry research evidence tools for DeepSeek Harness

Registers a janitor_scan tool that returns a deterministic, reviewable cleanup plan for session debris (plan docs, scratch scripts) with per-file verdicts and evidence; scan-only, never deletes.

Conflict diagnosis and repair for DeepSeek Harness plugins. Builds a capability graph of the running Loader profile and of the on-disk cordis.patch.yml layers, comparing imports, peer ranges and config against the host that is ACTUALLY running rather than the profile's possibly-stale copies, and reports the conflicts that fail silently: a fiber stuck PENDING on a service no plugin provides, a browser half injecting a client module nothing provides, a tool id two plugins both register, duplicate Loader entry ids, a row disabled by one layer and re-inserted by another, an import of a host subpath the running host does not export, and a cycle among inject declarations. Every finding carries its evidence, a confidence grade for who owns a name, and a repair with a risk grade — verify it with a dry-run difference table, or take the individual .patch.yml files that --fix-out writes (never auto-loaded, so picking wrong breaks nothing).

Read-only clinic for DSH session logs: 22 named diagnostics (token waste, retry and failure loops, context bloat, cross-session amnesia, idle MCP servers and skills), each reported with a replayable evidence locator such as session.v4.jsonl.zstd#L101-L110, and each finding turned into an executable prescription — a profile-patch YAML snippet or a command with current to suggested values, risk, verification and source — plus a feedback command that retires a prescription you reject, after which the report states it was removed by your feedback. Zero dependencies, default zero egress, session files are never written.