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/continuedev/continue/devbox-entrypointgit clone --depth 1 https://github.com/continuedev/continueWhat 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 | $0.00000 | $0.00262 |
| Opus 5 | $0.00000 | $0.00131 |
| Sonnet 5 | $0.00000 | $0.00052 |
| Haiku 4.5 | $0.00000 | $0.00026 |
Grade A, and why
devbox-entrypoint 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
Devbox entrypoint behavior (cn serve)
Context: runloop resumes a devbox by re-running the same entrypoint script, which invokes cn serve --id <agentId> .... Because the entrypoint always replays, the CLI must avoid duplicating state on restart.
- Session reuse:
servenow callsloadOrCreateSessionByIdwhen--idis provided so the same session file is reused instead of generating a new UUID. This keeps chat history intact across suspend/resume. - Skip replaying the initial prompt:
shouldQueueInitialPromptchecks existing history and only queues the initial prompt when there are no non-system messages. This prevents the first prompt from being resent when a suspended devbox restarts. - Environment persistence: The devbox entrypoint (control-plane) writes all env vars to
~/.continue/devbox-envand sources it beforecn serve, so keys survive suspend/resume. The CLI assumes env is already present.
Operational notes:
- Changing the entrypoint is expensive; prefer adapting CLI/session behavior as above.
- When testing suspend/resume, confirm a single session file under
~/.continue/sessionsfor the agent id and that follow-up messages append normally without replaying the first prompt.
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 · 13 lines · 0 tokens per session scan A 02e41663d793
devbox-entrypoint is a command published in the GitHub repository continuedev/continue (35,702 stars, last pushed 2d ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 262 tokens. 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
deslop
Check the diff against main, and remove all AI generated slop introduced in this branch.
MIGRATE_DESIGN
Design doc for the migration tool PR. Author: Sol ([email protected]). Co-authored-by: wakesync.
commit
智能生成 Git 提交信息并提交.
requirement-review
需求文档多角色评审（requirement-review）：需求文档 → 7-Agent 并行评审 → 重构高质量需求文档（Runtime 受控流程，0-7 阶段状态机）.
bug-check
Run automated tests and build checks first, then agent code review. For each bug found, propose or document a regression test.
release
Create a new release for the Starpod workspace.