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 skills add HazenSun/dsh-codex-bridge --skill delegate-to-dshgit clone --depth 1 https://github.com/HazenSun/dsh-codex-bridgeWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/hazensun/dsh-codex-bridge/delegate-to-dsh)<a href="https://agentmods.dev/skills/hazensun/dsh-codex-bridge/delegate-to-dsh"><img src="https://agentmods.dev/badge/skills/hazensun/dsh-codex-bridge/delegate-to-dsh/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/hazensun/dsh-codex-bridge/delegate-to-dsh"><img src="https://agentmods.dev/badge/skills/hazensun/dsh-codex-bridge/delegate-to-dsh.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What 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 5.1 | $0.00076 | $0.00738 |
| Opus 5 | $0.00038 | $0.00369 |
| Sonnet 5 | $0.00015 | $0.00148 |
| Haiku 4.5 | $0.00008 | $0.00074 |
Grade A, and why
delegate-to-dsh 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 10d ago.
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.
How it starts
The opening of the file, as written. The whole thing — 46 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Delegate to DSH
Use the DSH Bridge as an external implementation worker. Codex remains the planner and reviewer. Treat a short request such as “使用 DSH 处理” as sufficient routing intent; do not require the user to name tools, profiles, or an agent count.
Choose the delegation shape
Default to automatic routing. Before delegate_task, resolve the request to single or multi using delegation routing. Honor an explicit user choice of single or multi Agent. When the user leaves the choice open, prefer one DSH Agent unless at least two workstreams are meaningfully independent or an independent review materially reduces risk.
Record a concise rationale and any requested roles in the delegation request. Never claim a multi-Agent run merely because the prompt asked for one: verify the result's delegation evidence and warnings.
Before delegation
- Call
get_setup_status. If setup is notready, follow the bundledsetup-dsh-bridgeworkflow instead of returning a generic missing-configuration error. - Call
list_profilesand select profiles by their descriptions and declared limits. Never invent a profile or expose provider credentials. - Delegate only a bounded objective with observable acceptance criteria. Keep strategic decisions, secret handling, deployment, merge and destructive operations in Codex.
- Prefer Direct Mode. Spawn the optional
dsh_orchestratorCodex agent only when the user explicitly wants a native Sub-Agent thread.
Task lifecycle
- Call
delegate_taskwithprotocol_version: bridge.dsh.dev/v1alpha1, the configuredproject_id, selectedprofile_id, objective, acceptance criteria, and the resolved delegation decision. - The tool returns quickly. Use
wait_taskwith a bounded timeout orget_task; do not busy-loop. - When the task is terminal, call
get_task_result. - Review the summary, changed-file list, warnings and tests. Read the patch with
read_task_artifactin bounded pages when needed. - If evidence is insufficient or the patch needs correction, call
continue_taskwith precise review feedback. The same DSH session and worktree are reused. - Call
cancel_taskif the objective becomes obsolete or unsafe. Never claim cancellation has converged until the returned state is terminal.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 10d ago First seen · 46 lines · 76 tokens per session scan A 593c380534f6
delegate-to-dsh is a skill published in the GitHub repository HazenSun/dsh-codex-bridge (1 stars, last pushed 6d ago), licensed Apache-2.0. It adds 76 tokens to every session and 738 once invoked, about $0.0004 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 2026-08-31.
Other skills, from other repositories
shortfilm-prompt
A five-part method for writing prompts for AI-generated videos, such as Sora or Seedance. It organizes each prompt into the subject, setting, visual style, camera movement, and timed shots with sound.
ao-workflow-runner
A workflow engine that reads YAML files and runs tasks assigned to different named roles. YAML is a text format for describing structured settings and steps.
skill
Ten native DSH tools give a text-only agent eyes. Use these structured tools directly; do not shell out to the bundled Python scripts or reproduce their implementation. Vision API credentials and model settings are managed by the plugin, so tool calls do not receive credentials.
bcs-coordination
A coordination engine for tasks that involve multiple bots, including shared context, bot discovery, routing, communication, and workflow coordination. It also describes interactive uses such as games, role-play, brainstorming, and interview practice.
bbs-relay-pickup
A coordination procedure for agents that share work on tasks through a BBS task system, where BBS is the service used to claim, run, and report task steps.
bcs-coordination
A coordination system for several software agents, meaning separate automated assistants, working together.