docs: tool descriptions state the implemented contract; 2026-09-02 benchmark row - #125
Merged
Merged
Conversation
The MCP catalog and the LangChain toolkit told the model one line per tool and nothing about timeouts, ring buffers, atomicity, or when to pick exec over spawn; twelve sandbox_id parameters had no description at all. Every description now carries what the SDK documents, the docs/mcp.md table follows. Prompt-audit items F3 and F4.
The catalog said every tool takes sandbox_id, an exact 256 KiB tail, a fixed 1-16 fork range, checkpoints kept until deleted, permanent fleet-wide deletion, wake on any call, and idempotent release; the code scopes sandbox_id to sandbox tools, keeps whole chunks up to 256 KiB, caps forks at max_fork_count, expires checkpoints under the TTL, deletes the node's copy, wakes on guest access, and forgets a released id. The benchmark note now cites the A/B rows it rests on.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Two commits.
mcp/tools.go) and the LangChain toolkit told the model one line per tool and nothing about the 5-minute exec cap, the 256 KiB per-process ring buffer, write atomicity, the exec-vs-spawn choice, or what fork/checkpoint/promote guarantee; twelvesandbox_idparameters had no description at all. Every description now states what the SDK documents (prompt-audit items F3 and F4);docs/mcp.mdfollows. Tests assert tool names, not description text.Gates:
go build/go test ./mcp, ruff on the LangChain package.