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/madciapka/gemini-claude-plugin/executegit clone --depth 1 https://github.com/madciapka/gemini-claude-pluginWhat 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 | $0.00018 | $0.00460 |
| Opus 5 | $0.00009 | $0.00230 |
| Sonnet 5 | $0.00004 | $0.00092 |
| Haiku 4.5 | $0.00002 | $0.00046 |
Grade A, and why
execute 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 3d 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.
What it actually says
Use this when you have a planned subtask and want Gemini to do the work end-to-end: implement, verify, and write a structured handoff. Gemini runs with --approval-mode yolo and --sandbox by default, on a child branch it creates itself.
For read-only diagnosis or a single-shot answer, use /gemini:rescue instead.
For passive observation of a long Gemini run, use /gemini:rescue-stream.
Route this request to the gemini:gemini-execute subagent.
Raw user request: $ARGUMENTS
Operating rules:
- The subagent is a thin forwarder. It runs
node "${CLAUDE_PLUGIN_ROOT}/scripts/gemini-companion.mjs" execute ...and surfaces the launch payload, streamed events, and final handoff verbatim. - Default execution mode is background+stream. Add
--waitto run in the foreground when the task is tightly bounded and you want the result inline. - The user's task text may be inline ("implement X using Y") or a plan-file reference (
@path/to/plan.md, optionally followed by extra context). The companion script handles both. - Do not preset
--model. Do not pass--yolo(forced) or--read-only(incompatible with executor mode). - After Gemini finishes, the agent reads the handoff file from
handoffPathand surfaces its contents. That handoff is the report — do not paraphrase or summarize it before showing the user. - If Gemini did not write the handoff file, surface that as a failure with the final lines of the log so the user can diagnose.
- Do not auto-merge the executor's branch. The orchestrator (you or the user) decides merge timing.
- If no task was supplied, ask what Gemini should execute (and whether they have a plan file).
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.
- 3d ago First seen · 28 lines · 18 tokens per session scan A c841518f7d24
execute is a command published in the GitHub repository madciapka/gemini-claude-plugin (8 stars, last pushed 4mo ago), licensed MIT. It adds 18 tokens to every session and 460 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 2026-08-31.
Other commands, from other repositories
git
Git operations with intelligent commit messages and workflow optimization.
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.