Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/cpj-dev/dsh-plugin-cc/setupgit clone --depth 1 https://github.com/cpj-dev/dsh-plugin-ccWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00015 | $0.00584 |
| Opus 5 | $0.00008 | $0.00292 |
| Sonnet 5 | $0.00003 | $0.00117 |
| Haiku 4.5 | $0.00002 | $0.00058 |
Grade A, and why
setup scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
Run the setup and show the user the resulting readiness report verbatim. First run installs the pinned npm CLI — use a long Bash timeout (10 minutes) or run it in the background, and tell the user it is installing:
node "${CLAUDE_PLUGIN_ROOT}/scripts/dsh-bridge.mjs" setup "$ARGUMENTS"
What it does, end to end (each step skipped when already done — rerunning is a no-op):
- Get dsh. Default: install
@deepseek-ai/dsh@<HARNESS_NPM_VERSION>into the plugin data directory (npm install --prefix), write a wrapper pinned to Node >= 22.19, persist it. A persisted source install (oldharnessCheckout/dshInstall: harness) is migrated to that pin.--harness <checkout-path>uses an already-built DeepSeek Harness checkout instead (the plugin does not runpnpm install/build:lib;packages/sdk/servermust exist). A later no-args setup migrates back to npm. Requirements: Node >= 22.19 (harness floor; the plugin itself needs only >= 20),npm(default path), andpnpmfor profile plugin add (corepack enablewhen missing). Do not follow npmlatest/next. - The
ccprofile for multi-turn sessions: dsh-base + the SDK JSON-RPC server. Default path:dsh plugin --profile cc add @deepseek-ai/dsh-sdk-jsonrpc-server@<pin>plus that package's published peerDependencies (a server-only add cannot resolve@deepseek-ai/dsh-sdk-protocolat boot).--harnesslink:-installs<checkout>/packages/sdk/server. Approvalnever, verified via--dump-config. One-shot commands (/dsh:review,/dsh:critique, fresh/dsh:run) work without the profile;--session,--resume, and/dsh:importneed it.
After setup, the only remaining prerequisite is a DEEPSEEK_API_KEY (env var, $DSH_HOME/.credentials.yaml, or .env) — the report's next steps say so when it is missing.
--mode minimal|standard|anchored-standard persists this machine's default agent mode. The built-in default is standard (full catalog from request #1); /dsh:setup --mode minimal locks two tools for the whole run; --mode anchored-standard is the two-phase filter. Per-run --mode still overrides.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 20 lines · 15 tokens per session scan A f6dd5e40e9b9
setup is a command published in the GitHub repository cpj-dev/dsh-plugin-cc (106 stars, last pushed 5d ago), licensed MIT. It adds 15 tokens to every session and 584 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
review
Run an adversarial review. Raw arguments: $ARGUMENTS.
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.