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 qwwiwi/dashi-plugin-claude-code --skill doctor-dashi-plugingit clone --depth 1 https://github.com/qwwiwi/dashi-plugin-claude-codeWrote 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/qwwiwi/dashi-plugin-claude-code/doctor-dashi-plugin)<a href="https://agentmods.dev/skills/qwwiwi/dashi-plugin-claude-code/doctor-dashi-plugin"><img src="https://agentmods.dev/badge/skills/qwwiwi/dashi-plugin-claude-code/doctor-dashi-plugin/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/qwwiwi/dashi-plugin-claude-code/doctor-dashi-plugin"><img src="https://agentmods.dev/badge/skills/qwwiwi/dashi-plugin-claude-code/doctor-dashi-plugin.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 | $0.00146 | $0.02114 |
| Opus 5 | $0.00073 | $0.01057 |
| Sonnet 5 | $0.00029 | $0.00423 |
| Haiku 4.5 | $0.00015 | $0.00211 |
Grade D, and why
doctor-dashi-plugin scanned grade D with 2 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 9d 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.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
--settings <plugin-dir>/.claude/settings.json \ Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
`sudo`/`rm -rf`, default tier, the unit carries `--permission-mode How it starts
The opening of the file, as written. The whole thing — 152 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Doctor: migrate from the gateway to the channel plugin
This skill helps an agent move from the old gateway (a Python daemon that
spawns a fresh claude -p session for every Telegram message) to the new
channel plugin (one long-running Claude session per agent), and diagnose what
went wrong if the migrated bot misbehaves.
Why migrate at all: the old gateway hits a permission-dialog hang without a
bypass flag, has no guarantee the final answer is sent, and — past 2026-06-15
— becomes expensive because per-message claude -p spawn moves to a separate
billing pool. The plugin keeps one live session inside the Max subscription.
The doctor is read-only. It never restarts a service, never writes config, never prints a secret. It tells you what is wrong and points at the fix; you (or the operator) apply it.
The one rule that prevents the worst outcome
Never restart your own channel/gateway from inside your own session. That
kills the only comms link mid-turn, and the supervisor relaunches you into a
loop. Apply config changes without restarting (the old process keeps running old
code), and let an external actor do the restart — the operator in their
terminal, or another service. This is non-negotiable; see
references/03-lessons-learned.md §3.
Workflow
1. Run the doctor
# On a systemd host this is usually ALL you need — the doctor finds the
# channel-*.service unit whose WorkingDirectory matches the plugin dir and
# infers --env / --session / the state config from it (autodetect):
bun skills/doctor-dashi-plugin/scripts/doctor.ts --plugin-dir <...>/plugin
# Explicit flags always win over autodetect; pass them when the layout is unusual:
bun skills/doctor-dashi-plugin/scripts/doctor.ts \
--plugin-dir <workspace>/.claude/dashi-plugin-claude-code/plugin \
--settings <plugin-dir>/.claude/settings.json \
--mcp <plugin-dir>/.mcp.json \
--settings-local <plugin-dir>/.claude/settings.local.json \
--env <channel.env path> \
--user <your numeric Telegram id> \
--session channel-<agent> \
--no-autodetect # opt out of systemd unit discovery
What ships with it
5 files 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.
- 9d ago First seen · 152 lines · 146 tokens per session scan D 5f0dd107ca2f
doctor-dashi-plugin is a skill published in the GitHub repository qwwiwi/dashi-plugin-claude-code (16 stars, last pushed 9d ago), licensed Apache-2.0. It adds 146 tokens to every session and 2,114 once invoked, about $0.0007 per session on Opus 5. A static security scan graded it D with 2 findings (reads agent configuration directories, recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
ashlr-cost-refactor
Token-efficient refactoring discipline for Claude Code. Diff-first before editing, structural edits for renames, find-all-refs then one multiedit per file, no re-reads between edits, one genome proposal at the end. Reduces refactor session cost 40–65%.
ashlr-search
Genome-aware search habits for Claude Code. Prefer ashlrgrep and ashlrglob over native scans; compose multi-term queries; trust genome summaries; orient once before searching. Reduces redundant reads and search token cost 40–70%.
ashlr-codex-doctor
Check Ashlr's Codex plugin packaging, MCP manifest, hooks, skills, and workspace allow-list. Use when Codex setup, plugin loading, or MCP registration looks wrong.
ashlr-codex-genome-refresh
Refresh or maintain Ashlr project genome state from Codex. Use after architectural changes, schema/routing changes, or substantial refactors.
ashlr-codex-parallel-explore
Efficient parallel exploration habits for Codex. Use when a task needs several independent codebase questions answered before implementation.
ashlr-codex-savings
Inspect Ashlr token savings from Codex using host-neutral CLI and MCP surfaces.