Skip to content

FINDINGS.md

Latest commit

 

History

History
78 lines (56 loc) · 3.74 KB

FINDINGS.md

File metadata and controls

78 lines (56 loc) · 3.74 KB
Raw

beam_mcp: FINDINGS

Source: Ultraviolet origin/main = 778accb82759d3def30b2e1410e1aeb573f76dac. Nothing is committed in this repository yet. main had no commits at start.

The seam is nominal: demonstrated red, in Ultraviolet, before any extraction

The owner named normalize_tool_name/1 as the demonstration. It is one, and here it is failing, run in a throwaway worktree at 778accb8 with _build/deps copied (§4b), the test file removed and the worktree deleted afterwards:

$ mix test apps/hacktui_agent/test/mcp_injection_defect_test.exs

  1) test tools/call on that same injected tool reaches dispatch
     apps/hacktui_agent/test/mcp_injection_defect_test.exs:24
     Assertion failed, no matching message after 200ms
     The process mailbox is empty.
     code: assert_receive {:dispatch, :novel_tool, _args, _opts}

2 tests, 1 failure
REAL_EXIT=2

Test 1 passed: tools/list advertised novel_tool from the injected catalog. Test 2 failed: tools/call on that same tool never reached dispatch.

The cause, at server.ex:

:56-57   tools = Enum.map(state.tool_catalog.all(), &tool_definition/1)   # honours injection
:144-146 valid_names = Enum.map(ToolCatalog.all(), &Atom.to_string(&1.name))  # ignores it

An injected catalog is honoured when advertising a tool and ignored when calling it. The existing suite passes only because FakeToolCatalog's two tools are both present in the real catalog, so the two lists happen to agree.

Commit 1 cannot be behaviour-preserving. Reported, not worked around.

normalize_tool_name/1 calls HacktuiAgent.MCP.ToolCatalog.all() directly, and ToolCatalog is on the owner's stay-behind list. So the package cannot compile without changing that line, and changing it is commit 2's fix. Commits 1 and 2 collapse.

The only alternative is to hardcode the six Ultraviolet tool names in the package so the defect survives verbatim, which the scope forbids outright: domain data in a generic module is the thing this extraction exists to remove.

Two other changes commit 1 cannot avoid, declared

  1. The new/1 defaults. dispatch: defaults to &Dispatch.safe_call/3 and tool_catalog: to ToolCatalog; both stay behind. They must become injected. No test changes: all four server tests pass tool_catalog:, and the three that omit dispatch: never reach a dispatch.
  2. @server_name "hacktui-hermes" (server.ex:11) is Ultraviolet branding compiled into a generic package, and server_test.exs:34 asserts it. Keeping it makes the package announce another product's name; changing it edits a test, which commit 1 forbids. It wants to be an injected option. Flagged, not touched.

stdio.ex travels without test coverage

mcp_stdio_framing_test.exs cannot travel: it drives the real bin/hacktui-mcp binary, computes repo_root() as the umbrella root, runs mix compile there in setup_all, and is tagged :mcp_e2e. Making it run here needs a package-local launcher: new work, not a move. So the transport ships with no in-package test. Recorded as a gap, not hidden.

NOTICE: the difference the owner asked me to report rather than reconcile

Ultraviolet's NOTICE at 778accb8 reads, in full on the copyright line:

Ultraviolet
Copyright 2026 Ayla Croft

It names one role. It does not mention Sudo Apt Holdings LLC, Script Kitty, the ORCID, or any separation of ownership from authorship, consistent with slice 14 being unstarted. beam_mcp's NOTICE is written from the owner's text and carries all three roles. The two files disagree, deliberately, and reconciling Ultraviolet's is slice 14's work, not mine.