dsh-calm
Emotional checkpoint and conflict-repair plugin for DeepSeek Harness that soothes, seals heated exchanges, and restores clean context.
Install
$ dsh plugin --profile web add 'https://codeload.github.com/wobenshiwomu/dsh-calm/tar.gz/v0.1.0'Plugin Overview & Capabilities
AI-assisted organization based on the public repository snapshot. The content must be grounded in source evidence and does not replace compatibility or security verification.
Key Capabilities
- Deterministic classification of incoming user messages into categories like abuse, noise, mixed, substantive, and distress
- Calm mode trigger with tool guarding that blocks write and execution tools while permitting read-only tools
- Context folding and sealing that shadows conflict exchanges from the model context while retaining complete audit logs
- Rolling window termination to safely halt the session if exchanges lack substantive instructions after a configured number of turns
- Controlled sealed-record inspection via the open_sealed tool with redaction by default
- Crash and restart recovery that reconstructs ongoing soothe states directly from logged messages
Useful For
- Preventing LLMs from entering defensive or hallucinated argument loops during user frustration
- Securing agent tool usage by blocking destructive write operations while an exchange is being de-escalated
- Recovering long-running workflows with a clean context after heated interruptions
Who It Fits
- Developers deploying interactive DeepSeek Harness agents to end users
- Engineers requiring automated session de-escalation and conversation audit logs
- DSH workspace administrators looking to prevent accidental file modifications during heated prompts
Documented Limitations
- Context recovery summaries currently rely on heuristic extraction rather than LLM-based summarization
- Does not handle workspace or file system rollbacks, requiring external tools like dsh-rewind
- Message classification is rule-based rather than driven by a dedicated classification model
- Sealing must adhere to protected-head constraints and will degrade to standard log appends if the fold range is rejected
DSH Compatibility
Version-specific runtime evidence collected by DSH Plugin. A missing result means we have not tested that combination yet.
Security Signals
Objective signals discovered from package metadata and source inspection. These are not a guarantee that a plugin is safe.
package.json declares 0 runtime, 3 development, 3 peer, and 0 optional dependencies.
package.json declares DSH bundle metadata.
GitHub reports the repository license as MIT.
A root package.json was captured and can be inspected by the registry.
Public GitHub source metadata is available for this registry snapshot.
Source & Registry Notes
Public provenance, Registry classification, and the latest source check for this entry, kept separate from runtime verification.
- Source repository
- wobenshiwomu/dsh-calm
- Registry source
- GitHub · dsh-plugin topic
- Registry classification
- Plugin
- Source checked
- c698f63 · 2026-10-20
This project is independently indexed from public source information. DSH Plugin is not affiliated with DeepSeek or the plugin author. Always check the author repository before installation.
Repository Activity
- GitHub stars
- 5GitHub stars
- Forks
- 1
- Open issues
- 0
- Last commit
- 2026-10-09
- Last release
- 2026-10-07
Related DSH Plugins
Ranked by overlapping capabilities, use cases, plugin type, categories, and DSH profile.
Persistent multi-model agent teams with dynamic DAG orchestration, run monitoring, and token insights for DeepSeek Harness.
View plugin →Penetration testing mode bundle for DeepSeek Harness with exploration graphs, asset tracking, and vulnerability reporting.
View plugin →Comprehensive long-term memory, self-evolution, skills, task management, and multi-agent dispatch bundle for DeepSeek Harness.
View plugin →A bundle combining a runtime injector and task-aware reasoning-mode router presets for DeepSeek Harness.
View plugin →