Hard-stop consecutive identical tool calls after a configurable streak so the agent cannot spin in place.
Install
# from GitHub (first run asks for allowBuilds approval — follow the hint, retry)
dsh plugin --profile web add github:173787247/dsh-repeat-stop
Any plugin you install runs third-party code with your own permissions — it can read your files, use your credentials, and reach the network, and tool approvals don’t sandbox it. GitHub-sourced plugins also run build scripts at install time — pnpm blocks those until you allow them, so an install can stop with ERR_PNPM_GIT_DEP_PREPARE_NOT_ALLOWED or ERR_PNPM_IGNORED_BUILDS; dsh prints the exact key to add under allowBuilds in your profile’s pnpm-workspace.yaml, and the install works on the next run. Allowing a build is a trust decision: only install sources you trust, and pin a commit (github:owner/repo#sha).
README
DeepSeek Harness plugin: hard-stop consecutive identical tool calls (same tool + same arguments).
Pairs with dsh-tool-budget. Part of dsh-wsl-kit.
Where it sits
Hard-stops a streak of identical tool calls. It does not cap a varied tool session; that is dsh-tool-budget.
flowchart LR
calls["identical tool calls"] --> guard["dsh-repeat-stop"] --> stop["hard stop"]
Suite diagram and version snapshot: dsh-wsl-kit. This plugin is 0.1.2 (daily). Do not copy that matrix into this README.
Compatibility
| Field | Value |
|---|---|
| Plugin | dsh-repeat-stop 0.1.2 |
| Minimum dsh | ≥ 0.1.2 (web UI one-shot ?token= on Windows relay :3081) |
| Latest verified | See dsh-wsl-kit Compatibility (currently 0.2.0-rc.2) — single source of truth for the suite |
| Kit set | daily (also in github / full; fetch+net also in llm) |
| Cloud Flash | Use model id deepseek-flash (V4.1 Flash) in ~/.dsh/settings.yaml / llm-deepseek — not configured by this plugin |
| Agent Teams | Upstream experimental; not required here |
Suite floor versions: kit check-plugin-versions.sh. Fault tree: TROUBLESHOOTING.md.
Why
Official repeat-tool-reminder only advises and never blocks. Agents can still spin on the same failing call. This plugin denies the next call after a streak.
Default: 6 identical calls may run; the 7th is blocked. Changing args/tool or a real user message resets the streak.
Install
dsh plugin --profile web add github:173787247/dsh-repeat-stop
Restart dsh web. No new tool appears. When it fires, Trajectory shows dsh-repeat-stop: blocked.
Config
- id: dsh-repeat-stop
name: dsh-repeat-stop
config:
enabled: true
threshold: 6
exclude:
- job_output
- job_list
- job_kill
# include: [] # if set, only these names (wildcards ok) are tracked
| Key | Default | Meaning |
|---|---|---|
enabled |
true |
Master switch |
threshold |
6 |
Allowed streak; next call blocked (integer ≥ 2) |
exclude |
job_* | Names that never count |
include |
(empty) | If set, only these names are tracked |
vs dsh-tool-budget
| Plugin | What it blocks |
|---|---|
dsh-repeat-stop |
Consecutive identical calls |
dsh-tool-budget |
Total tool calls in the whole session |
Test
npm test
License
MIT
Links
More in this category
toby-bridges/api-relay-audit★ 872
Runs local security audits of AI API relays and LLM proxies from DeepSeek Harness, producing Markdown reports for prompt injection, model substitution signals, tool-call rewriting, error leakage, stream integrity, and profile-gated Web3 risks.
SeaOf0/dsh-redteam-model★ 679
Authorized-security DSH collection: nine work modes (redteam coordinator, pentest, code audit, binary analysis, attack-defense, AV evasion, incident response, cloud security, CTF solving) and fifteen runtime plugins, managed from a settings page with one-click deploy, install, update and uninstall.
howmp/dsh-pentest★ 606
Authorized pentest mode for DeepSeek Harness — exploration chain, assets and findings with a Web view.
PerryLink/dsh-auto-review★ 235
Second-model auto-review on the approval answerer chain: a read-only reviewer subagent returns structured allow/deny verdicts with reasons, fail-closed by default.
NanmiCoder/dsh-auto-mode★ 164
Adds an Auto permission preset between Workspace Write and Full access: routine work stays in the official workspace-write sandbox while the current session model reviews escalation and destructive calls, granting one exact wider access once, asking when the intent is ambiguous, and denying critical paths.
PerryLink/dsh-permission-rules★ 120
Claude Code-style declarative permission rules: ordered allow/deny/ask YAML rules matching tool names, arguments, workspace paths, and agent identity on the tools/pre-execute waterfall, with full session-log audit, dry-run mode, and hot reload.
Community comments
Comments are public GitHub Discussions. Loading them connects to GitHub and Giscus; a GitHub account is required to post.