CABTA

Tool annotations

No tools have read-only/destructive annotations

Add readOnlyHint or destructiveHint annotations to every tool so hosts can warn users before invoking.

All four hints declared on every tool

77/77 tools missing one or more hints — investigate_ioc (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); analyze_email (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); analyze_file (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint), +74 more. OpenAI's directory rejects tools where any of the four hints are missing or non-boolean.

For every tool, set all four hints (readOnlyHint, destructiveHint, idempotentHint, openWorldHint) to explicit true/false values that match the handler’s actual behaviour.

Descriptions match behaviour

1 tool describes read intent but its handler mutates — email_security_check (line 364: r = subprocess.run()

Rename the tool, rewrite the description, or move the side-effect into a separate clearly-named tool.

Tool test coverage

Only 4/77 tools referenced in tests (5%)

Write tests that reference each tool by name so every tool has at least one test.

Shell command execution

8 child_process/subprocess calls in production code — runs shell commands (src/agent/playbook_engine.py:173, src/agent/playbook_engine.py:826, src/analyzers/string_extractor.py:287)

Prefer library functions over shell-outs. If you must shell out, ensure all inputs are properly escaped.

Secrets stay with their owner

6 secret/sensitive values flow into network calls (MOBSF_API_KEY → dynamic, MOBSF_API_KEY → dynamic)

Audit where credentials are sent. A NOTION_TOKEN should only reach api.notion.com — never a third-party host.

Tool description accuracy

email_security_check: description implies read-only but handler writes/deletes/executes

Update tool descriptions to accurately reflect all capabilities — especially write, delete, or execute operations.